Slashdot Mirror


Sun Bashes Linux on (IBM) Mainframes

dagbrown writes: "An article linked from Sun's front page, entitled "Linux on the mainframe: Not a good idea" by Shahin Khan, Sun's chief competitive officer, has the interesting theory that Linux on mainframes makes no sense because, among other things, the VM/Linux combo isn't a very good match. What do the folks on Slashdot think?"

513 comments

  1. It makes sense... by Anonymous Coward · · Score: 2, Funny

    Like Solaris on pc

    1. Re:It makes sense... by mAIsE · · Score: 0

      since when is everything is about m$ ??

    2. Re:It makes sense... by Anonymous Coward · · Score: 0

      I agree with you! You have enlightened me! I shall no longer use Linux. As soon as I click Submit I shall go to my local software retailer and buy Windows XP. God please forgive me for going so wrong for so long with Linux.

    3. Re:It makes sense... by FooBarWidget · · Score: 1

      I'm sure everybody will he very happen if you give them instructions on how to run XP on mainframes.
      Imagine Teletubby wallpapers on the mainframes' displays!
      The world will become so much better!

  2. Interesting read. by gmplague · · Score: 1, Redundant

    Looks like an interesting read to me. But it seems like it is logical for sun to do this. They're still trying with their fledgling solaris, I'm sure this kind of bashing goes on at every company.

    --
    __________________________________________
    Take comfort in your ignorance.
    Grandmaster Plague
    1. Re:Interesting read. by RazzleFrog · · Score: 2

      What do you mean by fledgling? Solaris dates back to 1991. I would hardly call that a fledgling operating system.

    2. Re:Interesting read. by Big+Jason · · Score: 1

      Fledgling? Solaris is more stable and scalable than Linux will ever be. Go back under your bridge, troll.

    3. Re:Interesting read. by York+the+Mysterious · · Score: 1

      Reffering to a fledgling OS I would say maybe OS X or BeOS (it's a strech, but it never caught on so it works)

      --

      Tim Smith - Ramblings from Nerd Land
    4. Re:Interesting read. by JordoCrouse · · Score: 1

      Solaris was introducted in 1991

      --
      Do you have Linux and a DotPal? Click here now!
    5. Re:Interesting read. by Anonymous Coward · · Score: 2, Interesting

      I like to call it FUD. In order to make sure your product sells, you need to market (kiss booty) to sell the product, plus create FUD about the competitions plans. Look at the Itanic CPU from Intel, touts 64 bit computing which DEC/Compaq has had with Alpha for over 9 years.

      You make money by shrewd marketing, not by building a quality product i.e. Yugo's inception, Microsoft, Sun, etc.

      But there are exceptions to that rule, build a quality product and market it "just good enough" like Cisco, and the market is yours. This idea failed with DEC/Compaq assuming the Alpha will sell itself.

    6. Re:Interesting read. by 2square · · Score: 1

      Yes but will it continue to be more so? (stable & scalable). If sun phases-out pure solaris and merges linux into solaris it seems to me that solaris stability/scalability will always be >= that for Linux. Will Sun make linux run better on Sparc than on intel?

    7. Re:Interesting read. by hisholiness · · Score: 1

      What do you mean back to 1991? SunOS dates back to 1982!

    8. Re:Interesting read. by RazzleFrog · · Score: 1

      But it wasn't called Solaris now was it? Either way my point is more than valid.

    9. Re:Interesting read. by Anonymous Coward · · Score: 0

      Fledgling in the sense of market penetration. Solaris is a good OS and very stable. The only reason I don't use it on my X86 boxes is that Linux is cheaper with a feature set that is comparable.

      I have to add that Linux is a little easier to use for me. Anything is easy to use if you take the time to master it. IBM is just promoting Linux as revenge for Microshaft punking out on OS/2. OS/2 from a stability standpoint was superior to Windows but the user interface was a little clunky and I don't think IBM promoted and sold it right.

      The cream of the crop does not always rise to the top.

    10. Re:Interesting read. by Anonymous Coward · · Score: 0

      Sun did several things before the rest of the industry, so I doubt it's reasonable to call their product "fledgling." First they switched to Sys-V. Then they hit 64-bit years before the rest of the game, and went through all the associated growing pains. They have better threading support than most. They have better SMP than most. It's funny to hear Linux advocates dis Solaris, seeing as how Solaris is where Linux wants to be.

    11. Re:Interesting read. by HalfFlat · · Score: 2, Insightful

      Solaris may be more scalable, but more stable? It depends very much on the hardware.

      To pull an example out of the air, I don't think I've ever seen a stable Solaris on the Sparc Ultra 5 (their 'cheap' IDE-based workstation.) I've also witnessed some really nasty wedging with LDAP authentication and panics on Sun Ultrasparc machines fighting it out with Sun-branded RAID arrays under load.

      Maybe the latest version of Solaris is completely unlike its predecessors, but it seems a little unlikely.

      Some versions of the Linux kernel, together with XFree, GNU software and other tools, have been exceptionally stable on certain combinations of x86-based PC hardware. Given that Sun control their own hardware, it seems unfair to criticise Linux's stability when compared with Solaris.

    12. Re:Interesting read. by Anonymous Coward · · Score: 0

      The name "Solaris" refers to the entire distribution. Solaris 8 (or 2.8 as some note it) is a distribution that consists of SunOS 5.8, and a bunch of other packages. So no, your point is not valid. SunOS (as included in Solaris) is _not_ a fledgling operating system.

    13. Re:Interesting read. by RazzleFrog · · Score: 1

      So let's review the thread asshole.

      Original Poster said - "They're still trying with their fledgling solaris"

      I said - "What do you mean by fledgling? Solaris dates back to 1991. I would hardly [translate to NOT] call that a fledgling operating system."

      hisholiness said - "What do you mean back to 1991? SunOS dates back to 1982!"

      I said - "But it wasn't called Solaris now was it? Either way my point is more than valid."

      Ignorant, Asshole, shithead AC (That's you) said - "So no, your point is not valid. SunOS (as included in Solaris) is _not_ a fledgling operating system."

      In summary, I said it is not a fledgling OS and you disagreed by saying that it is not a fledgling OS. You might be the dumbest person on Slashdot. Congratulations.

  3. Maybe not, by Cardhore · · Score: 2, Informative

    But free software on a mainframe isn't bad. Remeber, we also have such things as "FreeBSD", "OpenBSD"; also "NetBSD." Yes, they're new to me too.

    1. Re:Maybe not, by Anonymous Coward · · Score: 0, Troll

      FUCK OFF!

    2. Re:Maybe not, by Anonymous Coward · · Score: 0

      As soon as I see somebody Flame a legit post swearing their head off,(Was this person saying that the BSD's were better? No. They simply mentioned that they exist)I realize that the person posting it does not actually posess a legitimate opinion on the matter, and they are trying to intimidate people in to believing that they feel strongly on the subject, without actually putting forth any sort of actual supporting evidence. This just might work in an argument that you are having with some random idiot on the street, but trying to do the same thing on-line is purely idiotic. What, are you going to go "Kick his ass"? I bet he/she is really scared. I better watch out though, you just might me 1337 h4x0r. It's stupid posts like that(and replies to it like this) that remind me why I only read the main page of slashdot. So go back to your oh so important system administration(tending to the WindowsNT network that you have running in your parents basement) that gives you such strong emotional opinions on the matter of Free Unicies. Go back to pretending that you have something to offer this world, that people actually give a shit about you and your "opinions". Heh, you're funny.

    3. Re:Maybe not, by Anonymous Coward · · Score: 0

      Sun says Linux wasn't designed to run in a VM and therefore it's not a good idea to do that. The BSDs you listed don't even run in a VM!

  4. Really? by TheReverand · · Score: 2, Funny

    Actually, I thought Sun would be saying lots of nice things about their arch-rival, and complimenting them on their excellent choice of operating system.

    1. Re:Really? by vicviper · · Score: 0, Redundant

      Uhhh moderators, this appears to be sarcasm, not flamebait...

  5. Of Course not! by Dragonshed · · Score: 1, Informative

    Why would it be a good idea when you can buy Sun machines with Solaris, and get much more quality computing power for your dollar!

    All this is is Sun actively protecting their brand.

    my 0.025$

    1. Re:Of Course not! by JWW · · Score: 2, Funny

      Yeah I love their talk about utilization. Oh no i f you buy one of these IBM boxes you have to plan out what your peak utilization is!!!

      And if you buy some sun machines to do the same thing you have to .... ummmm ... plan out what your peak utilization is!!

      The add that the server can't dynamically create more utilization capacity (extra hardware) dynamically. If anyone out there were selling a box that could do that, I'd buy it in a heartbeat. First we'd need some good nanobots, or maybe a replicator.....

    2. Re:Of Course not! by Zelet · · Score: 0

      IBM does do this with some mainframe lines. They have extra processors and storage dormant... when the customer say... gets slashdotted... they can call IBM pay a little fee and get the extra hardware activated. Sorry I don't have any links, but I worked on the Linux/Zseries/S390 advertising for a while.

      --
      ...And when they came for me, there was no one left to speak out for me." - Martin Niemoeller (1892-1984)
    3. Re:Of Course not! by aminorex · · Score: 5, Insightful

      Well then, you should buy a starfire, because you
      can keep it running, slap in new processors, new
      memory, and then suck them into a running
      partition.

      It seems that most of the criticisms of Shahin
      Kahn's article are based on ignorance. It's a
      fair assessment of the liabilities of using
      mainframe hardware for typical modern web service
      applications. IBM tried to save the mainframe
      from declining market share in a very ingenious
      way, and Linux and IBM have benefited from it,
      but that doesn't mean that it is competetive
      with Sun's hardware offerings for the same
      application environments.

      Not all of Kahn's objections to VMs are valid,
      however. The robustness arguments are good, but
      the performance ones are short-sighted. While
      s/390 Linux may not be tuned today, you can be
      confidently assured that it will be soon -- even
      if IBM has to fork the kernel to do it.

      --
      -I like my women like I like my tea: green-
    4. Re:Of Course not! by Anonymous Coward · · Score: 0

      I'm on the s390 tech side, that is not the case now. (if it ever was)

    5. Re:Of Course not! by afidel · · Score: 3, Interesting

      The add that the server can't dynamically create more utilization capacity (extra hardware) dynamically.

      Actually IBM's regular mainframes can. When you buy one of the higher end zSeries servers you get a box fully populated with ram and cpu's. If your liscense is for something less than the max # of cpu's and you later need to add capacity all you do is call IBM and they happily take your money and dial into your mainframe, they set a couple registers in the controller board and viola near instant hardware upgrade.

      --
      There are 4 boxes to use in the defense of liberty: soap, ballot, jury, ammo. Use in that order. Starting now.
    6. Re:Of Course not! by Anonymous Coward · · Score: 0

      off topic put if I were to contribute "my 0.025$" would I actually have to cut the 3rd penny in half or do I just round it off to 0.3$? :-)

    7. Re:Of Course not! by mashedpotatoes · · Score: 1


      The add that the server can't dynamically create more utilization capacity (extra hardware) dynamically. If anyone out there were selling a box that could do that, I'd buy it in a heartbeat. First we'd need some good nanobots, or maybe a replicator.....

      Actually, he was comparing two alternatives
      1) a single IBM mainframe running multiple Linux images.
      2) a Linux server farm.

      In the second alternative, a load balancer distributes traffic evenly across all of the servers. To increase the capacity of your server farm, you can add a new server and reconfigure your load balancer. This can be done while the rest of the farm is still up and running.

      However, dynamically increasing the capacity of a single mainframe would be more difficult. (Maybe impossible, but I admit I don't know much about mainframes) How would you do that? Add more processors? More memory? Add a faster disk? All of these options would require some disruption. (Again, I am not a mainframe expert, but adding cpu/memory while the thing is running seems difficult to me. Disks can be hot-swapable of course...)

      Anyway, it seems to me that Linux server farms work pretty good. Why move everything onto one really expensive mainframe? There don't seem to be many benefits to justify the added expense.

    8. Re:Of Course not! by Pig+Hogger · · Score: 2
      Actually IBM's regular mainframes can. When you buy one of the higher end zSeries servers you get a box fully populated with ram and cpu's. If your liscense is for something less than the max # of cpu's and you later need to add capacity all you do is call IBM and they happily take your money and dial into your mainframe, they set a couple registers in the controller board and viola near instant hardware upgrade.
      That's so typical IBM. A holdover from the days where they had two line printers, one who could print twice as fast as the other, rented at (of course) twice the cost. When you needed to upgrade, a trained monkey was sent and all he did was flip around two pulleys between the paper-feed and the motor...
    9. Re:Of Course not! by Dazza · · Score: 1

      However, dynamically increasing the capacity of a single mainframe would be more difficult


      And you then go on to say you don't have a clue what you're talking about.


      The perfect /.'er ?

      --
      -- "I know that this is vitriol, no solution, spleen-venting, but I feel better having screamed, don't you ?"
    10. Re:Of Course not! by maxpublic · · Score: 1

      The mainframe can sometimes be more cost effective. Once you scale up to a certain size in your server farm the cost of floorspace, hardware, electricity, and technical support time actually outstrips that of comparable computing power in a mainframe. The initial invest on the mainframe is higher at the outset but you save money over time.

      The server farm is certainly the better option when you don't have that many machines. But when you start getting in the range of a hundred or more then it's time to buy a mainframe.

      Max

      --
      My god carries a hammer. Your god died nailed to a tree. Any questions?
    11. Re:Of Course not! by norton_I · · Score: 2

      Sun and HP are both doing this now, as well, at least with CPUs (I don't think the do it with memory).

      I know Sun had problems with the "hot add CPUs" idea -- people just didn't want to plug a cpu module into a million dollar machine while it was running and handling millions of dollars of transactions a day. Also, I know at least one case of someone trying to do that, and having the Sun technitian blow 3 CPU modules before giving up on it.

      Given that the marginal cost of a CPU is low compared to the development costs, it really makes sense to ship disabled CPUs and turn them on as the customer needs them.

    12. Re:Of Course not! by KernelHappy · · Score: 1

      I see your point about how this seems unfair, but lets look at the flip side. Your an employee that is doing more work and making more money for your boss and not seeing a cent of it, you'd be pissed too.

      Now think back to how computers infilitrated businesses. Most companies were computerized because it saved them money or enabled to make them more money. Often the sale of big iron was accompanied by service contracts and engineering consulting making the transaction look more like a partnership in a project than a traditional sale of goods. Selling them the rights to use only a portion of the machine was good for both parties, a company with less need for power didn't have to make the decision between buying hardware for todays needs and then retooling down the line or spending more money upfront for hardware they didn't need. The manufacturer probably saw two benefits, one they have hardware in the field not generating revenue that brings them a tax credit and they know that they are almost guaranteed a sale in the future (to unlock the extra hardware). This is a win-win situation for both parties when you really think about it.

      --
      -- Button up, your ignorance is showing
    13. Re:Of Course not! by rogueroo · · Score: 2, Interesting

      I'm on the s390 tech side as well, and I'm not AC, _and_ you're wrong. z900 processors, and perhaps earlier ones as well, have processors that lie dormant. Pay extra money to IBM and they will provide authorization codes to "turn on" that extra hardware.

    14. Re:Of Course not! by amblin · · Score: 1

      I take offense to your trained monkey comment. My dad was one of your so called "trained monkeys". He was highly trained and highly skilled in both the hardware and software of IBM's old iron. He could tear down to the bare metal and put back together a system with his eyes closed or install low level software patches(non trivial in them days), you name it.

    15. Re:Of Course not! by Anonymous Coward · · Score: 0

      Anyone who could/can reliably fix old hardware had a foundation based in math, logic, and electronics.....

      I didn't do mainframe work but as a younger lad supported hundreds of customers using S-100 based systems... The Brave ones running MP/M and Multi-user NorthStar DOS --- (the crowd goes wild)

      The early version of CP/M required the user (or vendor ;;) to write a tty driver in assembler and write it in the boot track --- just to get CP/M loaded.....

      A truely dedicated user..... when they knew truley knew the machine (like an old car)

    16. Re:Of Course not! by Anonymous Coward · · Score: 0

      Blah to that. My dad has worked for Big Blue for over 30 years. Being the progeny of that legacy you tend to learn a few things. Those were definitely trained monkeys in there. The skilled professionals kept IBM where it is today (in the lead). It's called smart business, you dumbass.
      And BTW - I work for SUN.

    17. Re:Of Course not! by Anonymous Coward · · Score: 0

      Another example of Sun marketing...the only system that Sun ships today that Dynamic System Domaining works on is their ancient E10K with a heavily patched Solaris!

      None of their StarFire line supports this feature today. They have made announcements that the 15K WILL support this feature, but have yet to deliver the software patches to make it happen. There is actually some fear that the feature will initially only be available in Solaris 9 (due later this year), and that a backport of these features to Solaris 8 may take even longer.

    18. Re:Of Course not! by Anonymous Coward · · Score: 0

      hmmm. does anybody have a keygen?

    19. Re:Of Course not! by Anonymous Coward · · Score: 0

      This is already happening.

      TrueBlue/64 is going to be arriving soon...it's a 64-bit Linux tuned for zSeries, pSeries, and iSeries.

  6. Sun is not Linux's friend by LordNimon · · Score: 3, Insightful
    From a Linux advocates point of view, there isn't much difference between Sun and Microsoft. Don't be fooled by the saying "My enemy's enemy is my friend", because it doesn't apply here.

    Besides, Sun will attack IBM at any chance it gets.

    --
    And the men who hold high places must be the ones who start
    To mold a new reality... closer to the heart
    1. Re:Sun is not Linux's friend by JabberWokky · · Score: 3, Insightful
      Yes it does - as long as the tactics that made MS "our" "enemy" aren't shared by the enemies enemy.

      In other words, FUD is bad, even if it's pro-Linux or pro-BSD. Embrace and Extend into open source is debatable. Closed source isn't so much the enemy, imo, as the crap that a few companies have pulled with it.

      --
      Evan

      --
      "$30 for the One True Ring. $10 each additional ring!" -- JRR "Bob" Tolkien
    2. Re:Sun is not Linux's friend by Anonymous Coward · · Score: 1, Insightful

      At some point, the Open Source/Free Software crowd is going to have to snuggle up to someone if we hope to win over corporate America, and Microsoft would be the wrong answer. They snuggle with no one. The IBMs and Suns of the world have the PR and marketing machine necessary to make this happen.

    3. Re:Sun is not Linux's friend by pmz · · Score: 5, Insightful

      I have thought hard about whether Linux was a real threat to Sun and whether Sun is a good thing for Linux.

      Well, I've come to the realization that Linux is not a threat to Sun. Instead, companies like Dell, HP, Compaq, and IBM are the real competition. What's the catch? They all compete on hardware implementations. They compete on prices and features. Would I still buy a Sun server with Linux? Yes, for the same reasons I prefer Sun servers with Solaris: the hardware has benefits beyond whatever OS happens to be running.

      Is Sun good for Linux? Yes, because Sun can provide an absolutely top-notch hardware platform on which to run Linux. All Linux needs are some hardware RAS support features and device drivers, which Sun is probably capable of providing, and better C-compilers for RISC architectures, which could be improvements to GCC or a port of Forte C to Linux.

      It is not Sun vs. Linux. I'm convinced of that. Rather, the Linux community should be asking "What can Sun do for us?" rather than "What does Sun have up its sleeve?" These same questions should be applied to all the first-class hardware vendors. The more hardware that Linux runs well on, the better it gets for Linux. It's win-win.

      What about Microsoft? Well, that's another war on another front over different principles. Sun is an ally in this war, unambiguously.

    4. Re:Sun is not Linux's friend by RFC959 · · Score: 2
      From a Linux advocates point of view, there isn't much difference between Sun and Microsoft.
      I'm not so sure about that. True, Sun would probably just as happily turn around and bash Linux if they thought that would profit them, but that's not all there is to it. A Solaris shop is probably more likely to support Linux at some point than a Windows shop is, it being a bit less of a switch. Also, it's in Linux suporters' best interest to support the underdog - the more of the market (especially the server market) that MS has, the more they'll control the underlying protocols. Keeping the market fragmented can help keep protocols open, which in turn benefits the underdogs.
    5. Re:Sun is not Linux's friend by Anonymous Coward · · Score: 0

      Running Linux on Sun hardware is a crime and abomination before God.

    6. Re:Sun is not Linux's friend by jazman_777 · · Score: 1
      It is not Sun vs. Linux. I'm convinced of that. Rather, the Linux community should be asking "What can Sun do for us?"


      Rather, "What can Sun do _to_ us?"

      --
      Slashdot: Failed Car Analogies. Amateur Lawyering. Anecdote Battles.
    7. Re:Sun is not Linux's friend by Drazi100 · · Score: 0

      not as big an abomination as x86 architechtures

    8. Re:Sun is not Linux's friend by endemic · · Score: 2, Informative

      Sun sure seemed to be Linux's friend about a week ago while they were flaunting Linux on their new x86 server line.

      Check out the Sun Broadens Support for Linux feature.

    9. Re:Sun is not Linux's friend by Derkec · · Score: 2

      Sun isn't eager for Linux to be successful on the mainframe, at least partly because it believes it has a superior product. You better believe that Sun does want Linux to kick butt in the x86 PC market though. If you have linux PCs you might go for the Sun server while if you have MS PCs you might go for a MS server. There's a huge difference between Sun and MS from a Linux point of view.

    10. Re:Sun is not Linux's friend by arunkv · · Score: 2, Informative
      From a Linux advocates point of view, there isn't much difference between Sun and Microsoft. Don't be fooled by the saying "My enemy's enemy is my friend", because it doesn't apply here.
      If you look at Sun's home page, the article just below the one bashing Linux on the mainframe is:
      Feature Story: Sun Broadens Support for Linux Aggressive new program expands the role of Linux on entry- level servers.
      Go figure!
    11. Re:Sun is not Linux's friend by norton_I · · Score: 2

      Which they mention in the article. Basically Sun wants to use linux on low end servers like web servers that can be clustered to very high density to meed demand, and use their Enterprise line for "mainframe like" tasks.

      To some extent he is right. I don't really think that buying a mainframe to replace 20 servers, even if they aren't the 1k machines Sun was talking about. However, if you get to 100 servers, or you have a mainframe with spare capacity used for other purposes, I think Linux on the s390 can be a good choice.

    12. Re:Sun is not Linux's friend by zap42hod · · Score: 1

      So what do they mean with this:

      http://www.sun.com/2002-0206/linux/

      ?

    13. Re:Sun is not Linux's friend by Anonymous Coward · · Score: 0

      The problem is that despite how nifty it looks to have 100+ processors running at a gazillion MHz, you have a hardware platform that isn't as scalable as others.

      If you take a look at benchmark comparisons, the processor with the best scalability in production servers is the Power family from IBM. It speaks volumes that a 24 x 600MHz pSeries can run circles around a 24 x 900MHz StarFire on SPECjbb2000.

    14. Re:Sun is not Linux's friend by pmz · · Score: 1

      Benchmarks are a mixed blessing. It's always nice when a benchmark favors your brand, but the tables are easily turned just by citing another benchmark. A newer jbb2000 result, for example.

      This is why benchmarks are so highly overrated. What seems to be true one day isn't the next.

      Fact is, the IBM pSeries compete in the same market as the Sun Starfire servers. In this arena, the benchmarks are just going to bounce around as each company edges forward. The SPEC historical data reflects this.

  7. Linux on anything is good. by Em+Emalb · · Score: 1

    Because the more machine types it's used on, the better it will become, and the more people will use it.

    Beautiful in principle, but we will see how the sheeple react.

    btw, what do you think the majority of /.ers are gonna say? Linux sucks? Come on ;)

    --
    Sent from your iPad.
    1. Re:Linux on anything is good. by Anonymous Coward · · Score: 0

      the more people who use toilet paper, the better it becomes! yes! let's start a toilet paper revolution!

    2. Re:Linux on anything is good. by db · · Score: 4, Insightful

      You've got to be kidding me.

      The economy and most business models are not a 100% research and development, not-quite-stable environment.

      Don't tell a large business "Well, it will get better the more people who use it". They'll spit in your face. They need to know what works, and what works now, and what will continue to work in the future.

      Right now, Solaris works. Linux-bigots will sit and say "Well Solaris doesnt provide useful GNU utilities and is a boar when it comes to performance!" Well, yes it is, but it's been around forever, and when Sun says they can make it work, they will MAKE it work. You can't sit around and play with something for awhile in a 100% production environment, and rely on tools which have a sketchy (in a business-model sense) support base. It just cant, and wont, happen.

      Just my $0.02.

    3. Re:Linux on anything is good. by RFC959 · · Score: 2
      They need to know what works, and what works now, and what will continue to work in the future.
      Aw, hell no. They need to know which vendor is going to take them out to play golf and ply them with meaningless phrases like "industry leader", and "world class", and "enterprise level", and which choice is going to sound best to upper management. "But FooCorp's BarServer is used by 99% of the Fortune 500! It's the only enterprise-level scalable solution, with support for industry standards! Why, Quux Inc does a zillion-dollar-a-year online business based on this platform!"

      I wish I were joking, but this really does seem to be the way it is at all too many places, not that I would have any direct experience with THAT *cough*ATGDynamo*cough*.

    4. Re:Linux on anything is good. by zorander · · Score: 1

      This corporate need for what works now is leaving linux in many ways behind. In developing a real time motion controller for linux, I've found that the real time support sucks. I've tried three different APIs and the result I've arrived at isn't pretty. One of the more promising APIs (KURT) reliably and repeatably dropped interrupts. A 5ms interupt would become a 10ms interrupt would become a 15ms interrupt and so on. Is this "what works" I'm sure I wouldn't be having this problem with QNX. Fortunately the company I work for is small enough that software cost per license is an issue. I ended up writing a mediocre solution using no realtime API because with deadlines there was no time to fix up an existing RT API for linux to fit our needs. Linux has been the source of oh so many headaches as we've tried to push this product out the door. Will it work? yes, but I can tell you that a well tested, well debugged, stable system would not have so many issues when it comes to getting work done. If we had gone with QNX and sucked up the cost, our development time would undoubtably be faster, and the product more reliable. Sorry, Linux isn't ready yet.

      Brian

    5. Re:Linux on anything is good. by chris_7d0h · · Score: 1
      Don't tell a large business "Well, it will get better the more people who use it". They'll spit in your face. They need to know what works, and what works now, and what will continue to work in the future.
      I agree with you here. Regardless if it's Solaris, AIX or OS/390. These stuff have been around for a long time and most all major companies have invested and used these OSs extensively with great success. Especially now, proposing something unproven will probably be like a slap in the face, due to a lot of "their" failed "web projects" (ie. hype investments).
      --
      In a society that believes in nothing, fear becomes the only agenda ~ Bill Durodié
    6. Re:Linux on anything is good. by frank_adrian314159 · · Score: 2
      when Sun says they can make it work, they will MAKE it work...

      Well, I tend to believe that a company whose mainstay for the last 70 or so years has been service can make their system work, too. And a bit better than some "fly-by-night" firm that seems to have a "solution of the week" fetish.

      --
      That is all.
  8. I disagree by Starship+Trooper · · Score: 4, Interesting

    Running Linux on a mainframe doesn't change the fact that you must still maintain an expensive, proprietary system, defeating the whole purpose of introducing open standards like Linux.

    Running Linux on an IBM mainframe doesn't defeat the entire purpose of using open standards like Linux. You still get the man years of free testing, free software, interoperability, and speed. Or rather, IBM gets them. And by tying software you can't charge for to hardware you can, IBM will have come up with a business model for selling Linux systems for incredible sums of money. Quite an ingenious plan - selling Free Software.

    Sun's just pissed they didn't think of it first.

    --
    Loneliness is a power that we possess to give or take away forever
    1. Re:I disagree by Anonymous Coward · · Score: 0

      "an expensive, proprietary system"

      Like a Sun workstation?

    2. Re:I disagree by Starship+Trooper · · Score: 2

      "an expensive, proprietary system" Like a Sun workstation?

      Kind of, but orders of magnitude more expensive. So expensive they come with a repairman, know what I mean?

      --
      Loneliness is a power that we possess to give or take away forever
    3. Re:I disagree by bigmouth_strikes · · Score: 1

      "...standards like Linux."

      I'm sure you mean "the single standardized Linux distribution."

      --
      Oh, I can't help quoting you because everything that you said rings true
    4. Re:I disagree by sbuckhopper · · Score: 2, Informative

      Say, like an Sun E10000 or Sun E15000 which costs a ton more than that IBM server quoted (for about the same power)...oh yeah, and Sun won't let you buy one unless you order a service contract...meaning, they come with a repair man.

      --
      "Everybody knows the moon's made of cheese," Wallace.
    5. Re:I disagree by meff · · Score: 3, Interesting

      True, Sun *isn't* too happy about IBM pushing the mainframes first.

      I am pretty sure, whatever kernel IBM chooses to use on their linux mainframes will be THOROUGHLY tested and rigerously tortured and beaten to death until they know exactly how it's going to act. There will probably be a guideline to read for do-it-yourself people on how to make the most stable kernel for the mainframe you have.

      Running linux on the mainframe has TONS of advantages, and they will become more pronounced as it gets more popular and more used.

      Sun, give me a break, we know you have a big mouth but scaring everyone doesn't work all the time.

    6. Re:I disagree by georgewilliamherbert · · Score: 1

      Say, like an Sun E10000 or Sun E15000 which costs a ton more than that IBM server quoted (for about the same power)...oh yeah, and Sun won't let you buy one unless you order a service contract... meaning, they come with a repair man.


      In those cases, when you're paying for a Mainframe, you're getting a mainframe. Also, fairly, with the IBM systems running native OSes, big SGI and HP boxes, etc.

      What Sun is pointing out is that the behaviour of the Linux-in-VM systems from IBM appears to be inferior of that of an equivalent cluster of standalone Linux servers, which costs immensely less than the S/390 box and is easier to maintain.

      At the time I write this, there are three testimonials from people who say they evaluated or ran these IBM Linux on Mainframe/VM systems posted in the overall commentary, and all of them are relatively negative. I have no personal experience with that, but going by both the theoretical issues and the posted testimonials, I have to lean towards Sun being right on this one.

      As a disclaimer, I'm a heavily Solaris/SPARC oriented user and admin, and have specified and built multi-E10K sites. I appreciate Beowulf clusters and similar large Linux/FreeBSD/etc clusters but haven't built one yet.
    7. Re:I disagree by JordoCrouse · · Score: 1

      Linux means kernel here, not distribution.
      In a situation like this, I think that the set of packages will be the minimal set of standard utilities, plus a few platform specific toys.

      --
      Do you have Linux and a DotPal? Click here now!
    8. Re:I disagree by rgmoore · · Score: 4, Interesting
      Running Linux on a mainframe doesn't change the fact that you must still maintain an expensive, proprietary system...

      As opposed to running on a Sun system, which lets you run on a ... oh wait a minute. Those alternatives from Sun are mostly expensive proprietary systems, too, aren't they?

      You still get the man years of free testing, free software, interoperability, and speed.

      You also get a system that lets you migrate from your existing Linux systems to the IBM system without having to learn the quirks of a different Unix variant. If you use a supported distribution, you don't even have to learn a new distro. And if/when you decide that IBM isn't where you want to be, you can switch to many other hardware platforms with nothing more than a recompile. Sounds like you still get the key freedoms that Free Software is supposed to provide.

      --

      There's no point in questioning authority if you aren't going to listen to the answers.

    9. Re:I disagree by Skapare · · Score: 4, Interesting

      Sun's pissed that they can't run multiple instances of an OS on their E15K systems. You might get Linux running on it, and Solaris is the champ OS on the big Sun machines ... but they are not virtual machine systems. IBM's hardware design lets them run multiple operating systems in parallel on one machine, and even dynamically share processor between them. And then with the VM operating system loaded, you can create multiple virtual machines and run Linux in each one. And VM is very efficient at that. I once ran 6 instances of VM inside itself, nested all the way down. Surely you've heard of the case where IBM tested running 41,000 instances of Linux under VM. It can do that, though that many seems rather pointless. It does let you partition off the resources so you can give each service function you need to run with its own virtual machine, and thus it's own Linux. And on the larger S/390 and zSeries machines, you can even run OS/390 (of MVS legacy) in those virtual machines, and mix/migrate between them all one a single mainframe piece of hardware.

      Now personally, I wouldn't do exactly that. That's an awfully big basket with an awful lot of eggs in it. IBM hardware is quite reliable, but not so reliable that you can depend on getting "five nines" on one single machine. Sometimes there are reasons to take the whole machine down. IBM comes from legacy enterprise worksystems and usually early morning Sunday can be scheduled for maintenance purposes. In these days of e-commerce, you don't have such luxury. If you want to be up all the time, you need redundancy, and that machine way back in the corner of the room where "all the servers are gone" isn't redundancy (virtual redundancy, maybe, but you need real redundancy). You need several machines.

      That said, there are pluses to IBM's approach as well. If you need to add another class of service, or partition users apart from each other because one needs to do stuff that needs root access? Give them their own Linux virtual machine.

      OTOH, well managed, rows and rows of racks with 40 1U servers in each one, running Linux or BSD or NT or W2K or whatever, can be just as effective, if not more so. You can put dual 1.X GHz CPUs in those 40 machines in one rack ... that's 80 CPUs. That's quite a lot. The IBM zSeries can certainly compete, as can the Sun E15K. But those are going to be physically big, and power hungry, machines, too. Take your pick. There's no simple best answer; certainly not for everyone. All this is about is marketing, anyway.

      --
      now we need to go OSS in diesel cars
    10. Re:I disagree by dasmegabyte · · Score: 2

      Sun did think of it. They're pushing Linux like crazy, with machines starting as low as $1000 and great software support. The dozen Cobalt machines I admin attest to that. It's the reason why there's a sun released JVM for Linux, promoting it to the lofty standards of NT x86, NT alpha and Solaris, whereas us BSD and Mac users need to wait until Apple, IBM or the Kaffe boys to get off their humps.

      --
      Hey freaks: now you're ju
    11. Re:I disagree by digitalunity · · Score: 2

      Most people don't have the technical knowledge to understand the differences between Sun and IBM hardware. I think they also don't understand the different types of application needs.

      The mainframes will always lead the pack in bandwidth and thus are very good working with multi-GB to multi-TB datasets. When you have a high computational need such as Computational Fluid Dynamics or molecular modeling, a cluster of workstations or a dedicated beowulf is going to perform far better than the mainframe. When you look at mainframes like the Cray T3E, they really do have it all; but you pay through the nose. To replicate the processor performance of a Cray with commodity products would require about 1/10th of the money.

      Basically, people need to not worry so much about the raw numbers and take a minute to think about what they're intended to be doing.

      --
      You can't legislate goodness. Let each to his own destiny, by will of his freely made choices.
    12. Re:I disagree by RFC959 · · Score: 3
      Sun's pissed that they can't run multiple instances of an OS on their E15K systems.
      They might disagree with you. I believe they call it "Dynamic System Domains". I don't think you can run multiple /different/ OSes on them, but I'd be really curious to know how many people care to do that IRL.
    13. Re:I disagree by phajek · · Score: 3, Interesting

      "Sun's pissed that they can't run multiple instances of an OS on their E15K systems. "

      Hello? You obviously have never used an E15K or even a E10K with *does* domains for over 5 years now. Want some companies who use this? Ebay for one done. And they would be crazy to use Linux on it.

    14. Re:I disagree by TheLastUser · · Score: 2, Interesting

      Sun enterprise servers, in fact, do support dynamic system domains which allow a system to be partitioned into several machines, but that's not why I am replying.

      I agree with most of what was said in the article, however, I would disagree with their statement about Linux, re: "This is Linux. It's designed for Intel.". Linux is highly adaptable, while I would agree that it doesn't scale up too well, to say that it is "designed for Intel" is a bit much. After all, Linux runs on a wider range of hardware than Solaris does.

      Solaris has a very good kernel, in my opinion, superior to Linux. I'm no kernel hacker, but from a user's point of view, I have yet to run into a root exploit bug in the Solaris kernel, it has never corrupted my filesystem, and it is configurable without recompilation. The OS also contains a lot of useful stuff that you don't get in Linux, auditd for instance (Linux versions don't work on SMP systems) and disksuite is really useful. But it also lacks a lot of stuff, I usually install massive amounts of GNU software onto a Solaris install, including tcpwrappers, gnu tar, make, and openssl/ssh.

      For a mid to high end server, Linux is really not a substitute for Solaris, any more than MySQL is a substitute for Oracle. At the low end Linux rocks, and any other choice is money down the drain.

    15. Re:I disagree by Lumpy · · Score: 2

      This is an excellent point. 3 racks of 40 servers clustered is cheaper and easier to maintain than any IBM mainframe. If I lose a processor card in an ibm mainframe it will cost more than $1900.00 to replace. while I can buy another 1u server for less than that. and I get the side benefit that I just got new hdd,enet and power supply in that node.

      I agree, if you have the money to spend, spend it on cluster nodes. if youhave an unlimited budget and need to impress someone? buy the big iron.

      --
      Do not look at laser with remaining good eye.
    16. Re:I disagree by Anonymous Coward · · Score: 0

      Well... If your running Veritas then sure. But I've corrupted UFS several times.

    17. Re:I disagree by LowellPorter · · Score: 1
      "...Quite an ingenious plan - selling Free Software... "


      Linux is Free as in Open Source not $$$

    18. Re:I disagree by Anonymous Coward · · Score: 0

      And by tying software you can't charge for to hardware you can, IBM will have come up with a business model for selling Linux systems for incredible sums of money. Quite an ingenious plan - selling Free Software. Sun's just pissed they didn't think of it first.
      Didn't SGI come up with a similar brilliant plan a couple of years ago? How's their stock doing?

    19. Re:I disagree by thammoud · · Score: 1

      I disagree with you. We used AIX for four years and the thing was so unreliable it was a joke. The tcp system was crap. On the other hand, Solaris is simply rock solid.

    20. Re:I disagree by kg4czo · · Score: 1

      No, but IBM's distro of Linux is supported.

    21. Re:I disagree by Vulture_ · · Score: 1
      Actually you can do that on any arch (certainly not only mainframes) using User Mode Linux, which is, as the name might suggest, a Linux kernel that runs self-contained in a process on a 'parent' Linux kernel (which, in theory, could itself be a User Mode Linux kernel running in yet another 'parent' kernel, and so on).

      You don't need a mainframe to give everybody their own little root.

      Did I mention User Mode Linux is arch-independent? This also means you can run it on Sun boxes. (And why on earth would you want to run Solaris when you can run Linux? ;)

      --

      The only way the typical /.er can pick up a chick is with a forklift. -- AC

    22. Re:I disagree by Tony-A · · Score: 2

      run multiple /different/ OSes
      Debugging the OS while the machine is being used for production.
      Console debugging is very expensive if you have to do it on the bare iron.

    23. Re:I disagree by Anonymous Coward · · Score: 0

      Sun's not pissed, they are SCARED. Linux has revitalized the mainframe market. IBM is selling a lot of these things and taking business away from Sun's big machines.

      Even Sun's argument that Linux should only be used in server farms is dangerous to them. IBM can sell those too.

      IBM can sell you a server of any scale, running almost any operating system, which means almost any application you might want. Sun can't.

      Sun can sell you Solaris running a single image on a big machine or Linux running a single image on a small machine. That's it.

    24. Re:I disagree by Anonymous Coward · · Score: 2, Interesting

      41,400? Old news. That was Test Plan Charlie.

      Test Plan Omega--VM running in Basic mode on a ZZ7 (12-processor G6)--did 97,943 images.

      No one, to my knowledge, has seen how far you can push it with 64-bit zLinux on a zSeries box.

      Interestingly, you *can* add horizontal capacity on the fly with Linux under VM, although I don't think anyone is doing it yet. Here's how. Let's say that your bottleneck is handling incoming SMTP, and that you have some instrumented way to measure this--say, something that emits an ARM correlator when a message is received and another when it's delivered, and that this information is being fed back into something on VM that determines the average elapsed time per message has exceeded your acceptable threshold.

      What you do then is clone another image and configure it in the role of mailserver (the basic IP and hostname configuration requires changing three lines in /etc/rc.config, if you're running SuSE; to enable mail services would require setting SMTP="yes" in the same file). (The way we've been doing this is with a configuration template that is a full install; /usr is shared as a read-only disk between virtual machines. Thus all software is present on each image, but only the services used in a particular role are enabled. The default state is with no services but ssh enabled.) This cloning and configuring process takes--in the places I'm doing it--about three minutes per image, of which about two is spent creating the user and copying the configuration disk, and another one in booting Linux far enough to run the rest of the configuration scripts, change settings, and reboot. If you've precreated host records for all of this ($GENERATE is your friend), you just need to script adding the new virtual machine as another host for the "incoming" alias in DNS.

      Your old machine had Y total virtual machines, of which X were providing mail service, so the proportion of total machine time spent servicing mail is X/Y (yes, this is oversimplified; not all your machines will run at the same priority). In the new scenario, the proportion of the machine used for mail is (X+1)/(Y+1), which is larger than X/Y as long as Y>X.

    25. Re:I disagree by norton_I · · Score: 2

      Really? Out of the dozens of people I have heard bitch about AIX, not one has said it was unstable.

      I have seen it used as a network server, a build machine, a graphics workstation (no joke!), a personal workstation used for scientific simulations, and a database server, all under relatively heavy load, and nobody has complained about it crashing.

      Plenty of other complaints, though. Usually it goes something like "Would you believe this shit!!! XYZ is prototyped as returning a char instead of a long in the C libraries. At least it is stable, though"

    26. Re:I disagree by norton_I · · Score: 2

      While it is true, I think it is not as stable as VM if one instance crashes. Also, while they tout the ability to dynamically reallocate processors, a friend of mine who tried it (on an E10K) said that if you tried to do it while running Sun Cluster (I think), the whole machine crashed. He claimed that this was not documented anywhere, though it is possible he just couldn't find it. In any case, having to choose between dynamic partitioning and automatic failover is kind of stupid.

    27. Re:I disagree by CMonk · · Score: 1

      But what you can't do with the cheap intel box is replace those processors and disks while the system is running transparent to the OS(s) running on that machine. If I were to choose, I whould go with the IBM solution. Fewer power requirements, no cable mess, adding or subtracking capacity (whether that be disks or procs or memory or psus) without bringing the anything offline and no need to rent all that rackspace. TCO and flexibility are in IBMs corner here -- if you are serious enough about your application(s) to fork out the dough.

    28. Re:I disagree by johnnyb · · Score: 2

      According to IBM, replacing your servers with their system can save you $250,000/yr in ELECTRICITY alone. Not to mention floorspace.

    29. Re:I disagree by ibbey · · Score: 2
    30. Re:I disagree by Wdomburg · · Score: 2

      >the mind boggles... do you really believe that
      >Mandrake or RedHat will install on the s/390?
      >
      >Get your shit straight.

      Before insulting other people, you might want to "get your shit straight" yourself.

      Redhat Linux for the S/390 was released December 2001.

      http://www.redhat.com/software/linux/s390/

      SuSE for the S/390 was released October 2000.

      http://www.suse.com/us/products/suse_business/sl es /sles_s390/

      TurboLinux for the S/390 was released January 2001.

      http://www.turbolinux.com/products/s390/

      Debian has a beta distribution for the S/390

      http://www.debian.org/ports/s390/

      Matt

    31. Re:I disagree by jeremythehunt · · Score: 1
      Sun went after IBM because it is pushing Linux harder than any other major player. Sun has lost big sales due to companies moving to Linux. They know this and know that IBM's Linux servers big and small are sucking up their market share.

      My company recently benchmarked an IBM x330 (I know completely different servers than this aricle) against a Sunfire 3800 running ASIC design jobs. The Linux machines were 200% faster and we could buy 60 for the price of 1 Sunfire fully loaded. Sun's top engineers came in to tweak things for us to make it competitive. They couldn't. That's a half million dollar deal that they know that IBM won and they lost.

      I'm sure were many other customers are seeing the bang for the buck advantage of Linux servers.

      But what does this matter to the z class? Now we are more likely to puchase a z class for high end since most code devoloped on the x class machines will compile easily on the z. More money out of Sun's pocket.

    32. Re:I disagree by p3d0 · · Score: 2
      IBM hardware is quite reliable, but not so reliable that you can depend on getting "five nines" on one single machine.
      Do you know that the "z" in "zSeries" stands for "zero downtime"? In theory, that beats the pants off "five nines".
      --
      Patrick Doyle
      I mod down every jackass who puts his moderation policy in his sig. Oh, wait a sec....
    33. Re:I disagree by skribble · · Score: 1

      Err... IBM has a fine JVM... In fact IBM leads the industry in Java Development (Not Sun... they just try to control it). Also Apples JVM is quite nice too.

      --
      --- Nothing To See Here ---
    34. Re:I disagree by killmenow · · Score: 2

      Those alternatives from Sun are mostly expensive proprietary systems, too, aren't they?
      They have been. But with Sun's recent announcements and leanings towards releasing x86-based Linux systems, they are just sewing the seeds of doubt to get people on their bandwagon.
    35. Re:I disagree by plankers · · Score: 4, Informative

      Older versions of AIX, namely AIX 4.1 and 4.2, weren't very good, standards-wise. A lot of this has been fixed. IBM's latest version of AIX, AIX 5L 5.1, includes a lot of stuff they got through Project Monterey. Project Monterey was the combined effort between SCO, IBM, and Intel. When they joined, IBM had access to the SysV source for the first time, and started making their OSes more SysV-compliant. This was coupled with a general rewriting of all the device drivers in the OS, so that the drivers were more robust and handled errors better. If you watch carefully, you can see the effect these efforts have had on AIX, even in AIX 4.3.3. AIX 4.3.3 has had some interesting, destabilizing bugs in the last year. They were patched rapidly (patched, but IBM Support always had a good workaround if you couldn't just reject the faulty software package and put the machine back the way it was -- AIX has excellent package management!). Looking at the bugs, though, they were very low-level bugs that seemed to get exposed by development work throughout the OS. Given the sorts of things that are changing in AIX, it is actually a good sign that deep-seated problems are boiling to the surface, because that means that the developers are really doing their job.

      The other thing that seems to have improved AIX 4.3.3 a lot is the benchmarking. Dueling with Sun isn't a bad thing all the time. AIX 4.3.3, with the latest patches applied, has an excellent, capable, and very tunable TCP stack. It has also had a number of features backported to it from AIX 5.1. AIX 5L 5.1 has a lot more of the cool TCP features in it, and is even with Solaris 8 in those regards.

      In the past, compiling has seriously sucked under AIX. However, the IBM VisualAge C++ compilers for AIX are cool. They will compile just about anything, and if you cannot compile it then the programmer should probably go back and adhere to the various C/C++ standards. Generally I have the same trouble compiling certain software under either Solaris or AIX. Sometimes I even break down and install the GNU compiler collection...

      Right now, IBM has two machines that support virtual machines, or in mainframe/IBM terminology, logical partitions (LPARS). LPARs are supported in hardware on the pSeries 690 running AIX, and on IBM zSeries running z/OS or z/VM. LPARs are hardware-based, and if a processor, memory chip, or other system component dies it is able to prolong an outage by deallocating it. Currently AIX is not able to dynamically add resources to an LPAR without rebooting the LPAR, but that is coming. Otherwise, LPARs are totally independent instances of the OS. And the reboot to add resources takes an LPAR about thirty seconds (yes, 30). The LPARs do share a few things, like the system clock, and a management workstation for console access, but they are isolated in hardware.

      On the pSeries 690, the boundaries of the LPARs are 1 processor/1 GB memory. So you have to allocate whole processors and whole gigabytes of RAM to an LPAR. While this sounds like it's not such a great idea (and it isn't the best), it's okay. Any more granularity can be handled by the built-in AIX Workload Manager, which is able to manage OS-level resources like memory, processor time, disk I/Os, etc. in percentages based on users, groups, and process names. That isn't something Sun has, and often you don't need partitions on a machine, but instead just need to keep two pieces of software from fighting for resources (or need to cap a group of users to low memory or low CPU usage). Workload Manager is very handy in that regard. The IBM mainframes running z/OS or MVS can do timeslicing, where an LPAR can have 10% of all of the CPU time on the box, or 5% of RAM, etc. You can also create situations where you overcommit the resources on the machine but define priorities, to guarantee levels of service. So maybe you have three LPARs, and one can have up to 80% of the CPU if it's busy, and the other two can have up to 30% of the CPU, but LPAR 1 has priority (so it gets its 80% anytime it needs it). This is where the pSeries 690 is going, it just isn't there yet (IBM took all the guys that made MVS capable of this stuff and pointed them at AIX).

      I really know nothing about Sun's Dynamic System Domains. I do know it is similar in certain ways to the concept of LPARs, but isn't as flexible as the MVS/mainframe LPAR scheme. The OS instances are still isolated from each other in hardware. The two are probably nearly even when it comes down to it.

      If you consider the hardware these things run on you will see that a pSeries 690 with 32 CPUs equals a Sun Starfire with 72 CPUs. What IBM is doing in this regard is cool -- they are actively attempting to put fewer components into a machine. More components == more likelihood of failure, and therefore you need more redundant components. More components == more heat, which leads to more failures. It also means more electricity for more components. You can also cluster the big machines using a derivative of the IBM SP2 technology (you can also cluster smaller ones, like pSeries 660s, which are a hell of a deal, price/performance-wise). So while Sun's machines can go to 72 CPUs, all you get are extra components and extra heat (and extra service calls)

      IBM has been working with Red Hat for a while now. In fact, when Linux was ported to the S/390, there were two ports. I forget who the individual was that did one of the ports, but IBM had a team that did the other. There is an S/390 distribution of Red Hat Linux -- check any mirror site that is worthy (many don't carry it). It generally lags behind the i386 distribution, but only by a few weeks. Considering the market, hey, that's not bad at all. And IBM is doing a lot of work on Linux itself, from porting JFS to it, to adding a lot of what makes AIX a very scalable, very stable, very reliable OS. Check the IBM AlphaWorks site for examples. I am not sure what Sun is doing in that regards, or what Solaris can offer Linux as far as technology. That isn't an insult, it is just my unfamiliarity with Solaris.

      And finally, every machine is stable & reliable with a good system administration team that is knowledgable. I have been doing AIX admin work for about two years now, and I love it (having come from a Linux admin background). AIX has always had excellent filesystem support (true logical volume management built right in with real journalling filesystems, while you have to buy it from Veritas for Suns), and that, for me, makes it very easy to work with. And if you work with an OS for a while you get to know how it does things. I cannot speak for the admins that deal with Suns. I don't work with Suns, but half of the machines where I work are Sun machines (the other half is AIX, with two mainframes, and a handful of Linux boxes). And those Sun sysadmins can make their machines as stable and reliable as my AIX boxes. So it just depends on what you're trying to do, what you already have, and often, how much it'll cost you, because more often than not the machines can do the exact same thing, except one will be thousands cheaper.

    36. Re:I disagree by Wdomburg · · Score: 2

      >ok, i'll bite, smart-ass

      Again with the epithets. This hardly gives confidence in the strentgh of your arguments.

      >Suppose that I want to migrate one x86 linux
      >server (for example Mandrake 7.2 based, which
      >runs Oracle 8i well) to a s/390 VM. Will it be
      >easy because there is MDK for the s/390? I doubt
      >it.

      Well, before I address that, I quote your original post:

      the mind boggles... do you really believe that
      Mandrake or RedHat will install on the s/390?

      The question has nothing to do with whether it would make anything easier, just whether it will install. Which it will.

      As for whether it would be make the port any easier? On Mandrake, no, as it has no S/390 port.

      However, Oracle, like most of the ISVs that have released software for Linux/390, support SuSE exclusively, as they had the first comprehensive commercially supported distribution for the S/390.

      Even if you're not using a commercially supported product such as Oracle, there is a distinct advantage to being able to utilize the same platform on the S/390 as your current Intel machines. The most obvious one is retraining, integration with current systems (e.g. patch management), and ease of porting (yes, some software needs to be "ported" between distributions, due to different versions of compilers, libraries, and directory layout).

      >SUSE's say it includes XFree (I wonder how X can >run on an s/390... but I may be talking my ass >here since the last MF I worked on was an old >3090)

      The same way it runs on headless servers - remotely, either on an application by application basis, or through XDM.

      >Turbolinux asked me to register to see more >stuff,so I ditched it. what kind of freak use >that anyway?

      Not for me to say, but they have a signifigant customer base.

      >Debian is beta, but it is promising. Debian
      >dudes are serious, but their attitude kinda
      >kills them in the corporate world. no
      >proprietary software? haha, dream on.

      Personally, I've not been happy with the Debian boxes I've had to administer. However, there are a signifigant number of companies running it.

      >Red Hat's look like vapourware to me.

      Most vapourware isn't available for download:

      ftp://ftp.redhat.com/pub/redhat/redhat-7.2-en/os /s 390/

      >So, don't be so touchy, you sissy. Any serious
      >production environment of linux on s/390 will
      >use an IBM's blessed distro, not that crap.

      I actually had a S/390 onsite for evaluation last year. Do you know what distribution IBM recommended? SuSE.

      Do you know why? IBM doesn't have a Linux distribution. They have a set of kernel patches available to support the S/390, and links to Red Hat, SuSE and Turbolinux.

      Matt

    37. Re:I disagree by Teutates · · Score: 0

      Sun released a JDK because their Cobalt customers demanded support that wasn't there before. Sun is jumping on the bandwagon fast with their "Cobalt OS"...that and the user interface are the reasons they bought Cobalt Networks.

      You won't find them helping too much with a SPARC port of Linux, but they release cheap x86 hardware to run their software.

    38. Re:I disagree by Anonymous Coward · · Score: 0

      What kind of "Real BOFH" runs his database server on Mandrake? Fah.

    39. Re:I disagree by dillon_rinker · · Score: 2

      Zero equals one for sufficiently large values of zero...

    40. Re:I disagree by merdark · · Score: 1

      Ironically, fault tolerance and redundancy are exactly what sun mainframes give you (that's hardware redundancy). Mainframes are complex beasts, and in reality Linux doesn't offer anything at all to the mainframe market. As for running multiple virtual machines, it's not such a big deal, SGI's Origin series with IRIX already does this. And since IRIX was designed with this in mind, the OS itself has special support for distributing the kernel's among the hardware components (granted thier implementation is not the best). The point being, mainframes need special support and special staff to maintain. Adding in Linux which was *not* designed for such things and doesn't have the architecture to support mainframe class hardware just doesn't make sense. This is all Sun is saying. I really don't think that Sun cares if they can run multiple OS's on their mainframes. Solaris has years of design and support poured into it, and is made specially to run on Sun hardware. You can't get that kind of reliability and technical support anywhere else.

    41. Re:I disagree by Roxy · · Score: 1
      Older versions of AIX, namely AIX 4.1 and 4.2, weren't very good, standards-wise.

      I'm sorry, but I have to take objection to this statement, as I was running the X/Open, OSF, and NIST test-suites in the Bull/IBM cooperation for the AIX 4.1. It passed every test suite we run on it (with the mandatory exceptions when the test case was disjoint from the interpretation of the standard). In addition, IBM was running other test suites and passing them.

      You may challenge AIX 4.x on the grounds that it didn't comply with SunOS/Solaris, BSD, SysV.4 etc., but it did comply with SysV.2, X/Open, POSIX, NIST, OSF (with some exceptions).

      In addition, it has always had its own Administration system (remember LiSA anyone). and a number of quirks as soon as the standards were out of the way (like JFS, LVM, etc.).

      Critisize it all you want (there is lots to critisize), but don't misrepresent it.

      Roland B.

      --
      -- Roland Buresund MBA, MCMI, CISSP
    42. Re:I disagree by Anonymous Coward · · Score: 0

      but you can. Modern beowoulf systems can have nodes come on and off ad-hoc. Hot swappable nodes if you will. and while I do agree with the power requirements are high if compared to rack servers from just 2 years ago.. Most 1U servers now only draw from 100-150 watts. Espically the energy star rated units. Pentium4's draw nearly 1/3d the power the P-III chip did. along with hard drives are now drawing less power.

      Cable mess. true... but that is invisible as long as the rack doors are closed.. run the Big Iron with all the panels off.. there's a wire mess. (and a beautiful thing in my opinion.)

    43. Re:I disagree by phajek · · Score: 1

      I have been trained on it and I can state that the isonation between domains works very well on an E10K. Clustering and E10K work well together.

    44. Re:I disagree by hetz · · Score: 1

      But what would you do with z900?

      Computatiational jobs? z900 doesn't fit in that case, no matter what you'll do - it's great in I/O - not in computations..

      Of course - putting few cheap SMP boxes could help a lot...

      --
      nah, no sig... move on..
  9. I actually worked on Linux on a Mainframe... by Dixie_Flatline · · Score: 4, Informative

    And it was a serious pain in the ass. There were problems with the virtual machine suddenly giving up the ghost from underneath us, and we'd see Samba processes go wild for no reason whatsoever. We had load averages spike well into the hundreds, and it was like we were always scrambling to keep it running, as opposed to setting it up and just having it work. We used to tell the students that the machine had caught on fire and had (literally) fallen over. We were even thinking of doing up artwork.

    All those impressive demos where they have 32 hojillion instances of linux running on a mainframe are meaningless. Sure, you can do it, but it doesn't do anything. If you try actually working with the setup, you'll be rebooting your machine 10 times a day, and those mothers take forever to freakin' reboot.

    1. Re:I actually worked on Linux on a Mainframe... by Anonymous Coward · · Score: 0

      Nice try. What is your job title at Sun? Anyone who has tried to use Linux on a Mainframe knows it works suite.

    2. Re:I actually worked on Linux on a Mainframe... by Spamuel · · Score: 2

      So if he works at Sun his opinion is automatically suspect, is that what your saying? Wouldn't that also make your opinion suspect because you're a Linux user?

    3. Re:I actually worked on Linux on a Mainframe... by Anonymous Coward · · Score: 0

      Lets see, students and mainframe from IBM....yea sure and I have some swamp land in CA that Linus Trovalds used to own to sell you. You sir are speaking out your arse.

    4. Re:I actually worked on Linux on a Mainframe... by darkonc · · Score: 4, Funny
      We used to tell the students that the machine had caught on fire and had (literally) fallen over.

      I gotta hunt down and re-digitize the video that a set of students did back in 1992. At that time the UBC Computer Science GraFic lab had a stack of IBM RS-6000 computers running AIX (which I sometimes pronounced 'aches') which tended to crash far too often.

      For their animation lab one group's video was an RS6000 shaking, smoking and melting down into a grave with (computerized) flowers sticking out of it and a gravestone blinking 888 (which the 3-digit LED display would do after a general system panic)

      With later versions of AIX and some coddling from your's truely, I was able to make the IBMs a good bit more stable -- but never quite as rock-solid as the SGIs running IRIX -- and they never really outlived their reputation.

      Then, of course, there was the obligatory IBM PC... Let's just say it was really good for running Castle Wolfenstein (the original).

      --
      Sometimes boldness is in fashion. Sometimes only the brave will be bold.
    5. Re:I actually worked on Linux on a Mainframe... by Pig+Hogger · · Score: 2
      ...
      and those mothers take forever to freakin' reboot.
      Reboot? Do you mean IPL????
    6. Re:I actually worked on Linux on a Mainframe... by Anonymous Coward · · Score: 0

      Your DUH! factor is quite high. If he worked on Linux on an IBM Mainframe, he probably wasn't working at Sun.

      I'd explain further, but I it would spoil all the fun for you as you explore the limits of logic.

    7. Re:I actually worked on Linux on a Mainframe... by digitalunity · · Score: 2

      Not necessarily. It only shows a little about his motivations for writing the piece. There are always alterior motives, right?

      --
      You can't legislate goodness. Let each to his own destiny, by will of his freely made choices.
    8. Re:I actually worked on Linux on a Mainframe... by Anonymous Coward · · Score: 0

      He probably means IML and IPL!

    9. Re:I actually worked on Linux on a Mainframe... by Lando · · Score: 2

      Hmmm,
      The original Castle Wolfenstein was only released on the Apple II. Perhaps you mean Castle Wolfenstein 3D?

      Lando

      --
      /* TODO: Spawn child process, interest child in technology, have child write a new sig */
    10. Re:I actually worked on Linux on a Mainframe... by killmenow · · Score: 2

      OK. I know it's off-topic, but if the original Castle Wolfenstein was only released on the Apple ][ platform, then how was I able to play it on my Atari? (And no, it's not because they were both powered by the Motorola 6502.)

      Perhaps it's because the Apple ][ release came out in 1981, followed by the Atari release in 1982, the Commodore 64 in 1983, and DOS in 1984.

    11. Re:I actually worked on Linux on a Mainframe... by wobblie · · Score: 1

      I know, I bought one for the house, and it was a mess. Crashing constantly and a total pain in the ass. After the 294,768,345 th support call to IBM, and 46.8 million dollars later, i finally switch back to Win2K. Now everything's great!

    12. Re:I actually worked on Linux on a Mainframe... by rogueroo · · Score: 1

      An IML would reload _all_ partitions on that box. An IPL would reload only the partition in question. He probably meant just what he said.

    13. Re:I actually worked on Linux on a Mainframe... by fw3 · · Score: 1
      Well yeah that was like year 1 for aix 3 / rs6000. The rev1 compiler was awful - that got fixed by the ibm mainframe compiler group by '93.

      I've run numerical analysis on rs6000's for a decade now and put the boxes thru their paces. month-long computations with daily operational stuff still chugging along fine. Uptimes typically 6-12 months between reboots.

      I had no beefs with AIX for stability, and I've yet to see a linux on which I could pull out one failed (non-raid) disk, swap in another, allocate space, mount and continue operations all without a reboot. and that was back in '94.

      IRIX otoh oi! our indigo's used to crash when you looked at 'em crosseyed. I'm sure there were good things about them but their reliability was never my idea of rock-solid

      ymmv

      --
      Linux is Linux, if One need clarify their dist: <Dist>/GNU Linux
      bsds are of course just BSD
    14. Re:I actually worked on Linux on a Mainframe... by morcheeba · · Score: 1

      blinking 888 (which the 3-digit LED display would do after a general system panic)

      Ha ha! I would have made it blink 12:00, but that's just my sick sense of humor.

    15. Re:I actually worked on Linux on a Mainframe... by Lando · · Score: 1

      Nod,
      Looking it up on the web I notice that you are correct. I guess that by the time it was released on the 64 or DOS that I had tired of the game.

      I remember playing the game at the local ComputerLand and being somewhat desperate to find a copy that I could play at home. Never saw one though.

      Sorry about that...

      Lando

      --
      /* TODO: Spawn child process, interest child in technology, have child write a new sig */
  10. ...troll by Drunken_Jackass · · Score: 0, Troll

    Too bad you can't moderate a whole story...this is a troll if ever saw one!

    --
    There are 01 types of people in this world. Those that understand binary, and me.
  11. Are you kidding?! by Neck_of_the_Woods · · Score: 0, Troll


    Sometimes I think your just baiting people. This should be interesting.

    The next one will be, "Microsoft Claims it can server webpages 2x Faster on IIS3.0 than Apache!".

    Stand back, even water will not save you from the flames!

    --
    Neck_of_the_Woods
    #/usr/local/surf/glassy/overhead
    1. Re:Are you kidding?! by nomadic · · Score: 2

      Why do /. posters insist that everything they don't agree with MUST be a troll? You know, maybe he really believes what he's saying, and isn't just writing it in order to get a response from you personally.

    2. Re:Are you kidding?! by Neck_of_the_Woods · · Score: 1


      No I did not take this personally I was just saying stand back, people here are not going to like this. Honestly I could really care less as it has no effect on my world, I am just a gimp that has to run in my wheel of Microsoft to pay for my jeep, boat, and surfboards. I smile while I do also.

      My point, ahhhh well it was this whole story is going to look like one big troll. That was all, enjoy.

      --
      Neck_of_the_Woods
      #/usr/local/surf/glassy/overhead
    3. Re:Are you kidding?! by Ixe · · Score: 1

      hey would some of the nice /.ers mind bringing me up to speed on what all the mods mean like troll :) I'm new go easy, but I'm not the ordinary microsloth dope either, I'm running linux and I actually understand a good portion of it now. - Forgive me, I'm just trying to force myself to post for once instead of just reading all the stories.

      --
      Sigs pose an operational security risk and help the baddies aggregate data. I guess commenting does too, oops.
    4. Re:Are you kidding?! by Neck_of_the_Woods · · Score: 1

      Troll in relation to slashdot or a newsgroup, and I am just going on what I believe it to be:

      Posting something for one purpose, to piss people off and or "troll" for a response. IE like throwing bait of the back of a boat to get a fish to bite, hence the term.

      If someone is saying a post is a troll they are pretty much saying it is a waste of space, and only there to get the people angry and posting more on an emotional level then about what the topic is about.

      I typical troll here would be, "Microsoft is great, and linux is for bean headed slackers that care more about being cool in the eyes of thier peers than making money."

      I think that about sums it up, please someone correct me if this is a bit off.

      --
      Neck_of_the_Woods
      #/usr/local/surf/glassy/overhead
    5. Re:Are you kidding?! by Daengbo · · Score: 0

      Posting something for one purpose, to piss people off and or "troll" for a response. IE like throwing bait of the back of a boat to get a fish to bite, hence the term.

      You know, that's funny, because I always thought that's what it meant, but whenever I've seen the fishing term ,it's been spelled "trawling." Eventually, I came to think that was the origin, but that it was mis-spelled and took the meaning of troll, the ugly-under-the-bridge-with-the-billygoats-type. Just my thoughts. Dan
  12. What about independent testing by javacodewarrior · · Score: 2, Insightful

    Like the man said, this sounds like the normal responce from sun, I'm sure Microsoft should have simular arguments for why MS is better. Unless we get a complete third party to analysis with no aligence to any OS. And maybe on that day pigs will fly.

    --
    http://www.funwithpenguins.com
    1. Re:What about independent testing by aminorex · · Score: 1

      Your confidence is misplaced. The arguments
      would have to be very dissimilar. Microsoft
      doesn't make robust hardware for enterprise scale
      application services.

      Sun makes great hardware, for these kinds of apps,
      where robustness and scalability are crucial.
      It's a sucky horrible place to work, and most of
      their software is useless (Java excepted) but
      the hardware is way cool.

      Now if Sun would drop Solaris, those mainframe
      sales would be going to Linux on Sun hardware
      instead. But some clueless suit bought SVR5
      10 years ago and still can't admit that it was
      a big, expensive mistake.

      --
      -I like my women like I like my tea: green-
    2. Re:What about independent testing by Wdomburg · · Score: 2

      >But some clueless suit bought SVR5
      >10 years ago and still can't admit that it was
      >a big, expensive mistake.

      Actually, they bought SVR4; SVR5 wasn't introduced until March 1998, with the release of Unixware 7. To the best of my knowledge, noone has ever bothered to license it.

      Matt

  13. Activity and mud throwing may be a good thing. by suso · · Score: 3, Interesting

    This may sound a bit odd, but it could be that the mud throwing that Sun is doing could end up being A Good Thing(tm) for all Un*xes just because it bring s more media attention to our community. Sun isn't directly saying that Linux sucks or that it's worse than NT or whatever, they are drawling attention to the use of Linux on mainframes of all things. So the drawn out fact that Linux is being used on Mainframes and being acknowledged by two major companies could result in good juju.

    1. Re:Activity and mud throwing may be a good thing. by bilbobuggins · · Score: 1

      that would be nice, but unfortunately what most people (management types) are going to come away from this with is that a big established company they've heard of before is telling them not to believe the hype about something that so far they've mostly just heard as a buzzword.
      i don't see any way this could end up being good publicity.

  14. New title? by Ded+Bob · · Score: 2, Interesting

    Chief Competitive Officer? I have never heard this title before. Is it new?

    1. Re:New title? by Andrewkov · · Score: 2, Funny

      Yeah, his job is to sling FUD at the competition.

    2. Re:New title? by WolfWithoutAClause · · Score: 2

      Possibly.

      For those not familiar with Terry Pratchett, that would be something like a 'spin doctor'.

      For those familiar with Terry Pratchett; that would presumably be Sun speak for 'head liar'. ;-)

      Every company should have one. Actually, come to think of it, every company probably does have one ;-)

      --

      -WolfWithoutAClause

      "Gravity is only a theory, not a fact!"
    3. Re:New title? by Anonymous Coward · · Score: 0

      I believe this is a pretty common position, at least in tech companies. I know Palm has one for keeping track of Pocket PC.

    4. Re:New title? by 2Bits · · Score: 3, Funny

      Yeah, this is for people whom you don't know what to do with. You give them a nice title with no responsibility, until they figure out that they are useless and they leave. But some of them will never do.

      In our company, we have some many "chiefs", and not enough Indians. I'm working on a project, and I'm the only architect, system designer, programmer, QA, integration engineer. In another word, a one-man show. But I have to report weekly status to 5 chiefs, each one wants it in a different format.

      The 5 are:
      - Chief Technology Officer (boy, this is a high visibility project, even the CTO has to know the weekly status. This guy wants the status in MS Project format).
      - Chief Product Strategist (What the heck, the guy is responsible of another product, but I don't know why he wants the weekly status of my project. But he makes sure that the CEO knows that he MUST get the status every week, so I can't get the CEO (of the division, that is) off my back. This guy wants it in Excel format, as he has a lot of product matrices in Excel).
      - Chief Competitive Office (well, I never figured out his role, and will never figure out why he wants the status either. Ah, but he wants it in PowerPoint. He doesn't want to waste his time to recreate the information when he has to do presentation, does he?).
      - Chief Project Management Officer (he can't even keep up with other real projects, but he still run after me every week for this stupid pilot project/prototype. Another MS Project format guy.).
      - Chief Customer Service Officer (what the fuck, the project I'm working on is just a prototype, for god sake... This guy wants the status in his searchable database, so I just give him all files that I gave to the other 4 chiefs.)

      Life sucks sometimes.

    5. Re:New title? by Anonymous Coward · · Score: 0

      Just imagine a beowulf cluster of these guys...

    6. Re:New title? by RetroGeek · · Score: 1

      BC = beowulf cluster

      BC1: Hey let's do Web adds
      BC2: I know, we could put up banners
      BC3: Wait we could pop the adds up
      BC4: Even better we could run Flash

      Yeah right, like we need these guys talking to each other....

      --

      - - - - - - - - - - -
      I am a programmer. I am paid to produce syntax not grammar. Deal with it.
    7. Re:New title? by WolfWithoutAClause · · Score: 2

      Beowulf cluster? That's called a government isn't it?

      --

      -WolfWithoutAClause

      "Gravity is only a theory, not a fact!"
    8. Re:New title? by UncleFluffy · · Score: 1

      Beowulf cluster? That's called a government isn't it?

      No, but close. The phrase you're thinking of for "government" does indeed include "cluster", but at the beginning rather than the end.

      --

      What would Lemmy do?

    9. Re:New title? by chris_7d0h · · Score: 1

      Sounds like you're working for "///".

      --
      In a society that believes in nothing, fear becomes the only agenda ~ Bill Durodié
    10. Re:New title? by RicRoc · · Score: 1

      I know where you're comming from man, and I feel for you. I've been there myself until I gave them a good talk-to. It helps, as long as you are well-formulated. Pointy-haired bosses are not as air-headed as they seem! :-)

      I suggest you have a talk with the boss, and reason with him. Just remember to be polite and use the words he uses himself!

      --
      Who?
    11. Re:New title? by shani · · Score: 1

      what the fuck, the project I'm working on is just a prototype, for god sake...

      The only way this will be "just a prototype" and not get foisted on some unsuspecting customer is if you make the code self destruct in some way.

      My advice: do the full product lifecycle here, i.e. requirements, design, document, implement, test.

      Do you think your sales reps are going to understand that this is "just a prototype"?

  15. I think... by sterno · · Score: 1, Offtopic

    ...that to really make a fair comparison, I'll need access to a mainframe running Linux. So, if Sun would be so kind, please send me a mainframe so that I can check your conclusions. I promise that I'll write a very thorough article in exchange.

    --
    This sig has been temporarily disconnected or is no longer in service
  16. Solaris by Ashcrow · · Score: 1

    It seems that Sun is a bit scared about Solaris losing out to Linux. They are both Unix(-like), both solve the same problems but Linux does it cheaper and allows you to look ath the source code. While Sun has used Linux to gain money and new friends (penguin computing) they would rather do their own controlling from Solaris.

    1. Re:Solaris by Bungie · · Score: 2, Insightful

      They are both Unix(-like), both solve the same problems but Linux does it cheaper and allows you to look ath the source code.

      Actually, the price of Solaris is not really that expensive unless you are using a system with very many CPUs. A single user copy if you download it is pretty much free. It is the hardware needed to run it on that is expensive. Also, the solaris source code is available.

      I doubt Sun really cares that badly about the success of Solaris so much as they care about the failure if IBM. As far as I can tell, this article is mostly pushing the fact that a cluster of low end Sun boxes running Linux will be better than an IBM mainframe running Linux.

      --
      The clash of honour calls, to stand when others fall.
  17. Sun is selling Solaris by npietraniec · · Score: 1

    ...Plain and simple. Sun wants to sell solaris, and IBM is competing with them. This shouldn't suprise anybody... Plus, they have a point about the VM. 2.4 VM still needs work if you want to run it on a mainframe. Not taking $$ into account, Solaris would probably be a better choice.

    1. Re:Sun is selling Solaris by Ded+Bob · · Score: 1

      How much does Sun charge for Solaris on their mainframes (or are they called Enterprise Servers)?

    2. Re:Sun is selling Solaris by walt-sjc · · Score: 2

      No. Sun is selling HARDWARE. Sun has been resisting Linux because Linux allows you to use just about ANY hardware. Solaris only runs on Sparc (Yes, I Do know about Solaris x86. It's such a pile of crap that nobody uses it, therefore it might as well not exist.) Sun is worried that if all their software partners ported to Linux that they would lose sales. They are right to be worried.

      BTW, I DO like sun hardware, but depending on the needs of the application, I run various OS's on various hardware. If all the software that is Sun only would run on Linux, this would reduce the need for sun hardware quite a bit.

      BTW, Solaris is for the most part free. They do require you to buy a copy of the media for something like $90 (which I did, for both sparc and x86) but you can install it on as many systems as you like.

      Solaris is better that Linux for some tasks, but not many. Linux is pretty rock solid and has pretty decent performance. Don't forget the BSD's either, which seem to make better firewalls, proxies, etc. due to network performance and other issues.

    3. Re:Sun is selling Solaris by Anonymous Coward · · Score: 0

      8 CPUs or less, Solaris is free. As in the not freaky, hippy, revolutionary meaning of the word.

  18. And how many Solaris VM's can I run? by Halo- · · Score: 1

    Perhaps the point is that the user might not want to have to buy a seperate box for each of their customers. Maybe Sun wouldn't be so upset if they had a similiar technology. (Which they don't, to my knowledge... at least not to the same level of scalability)

    1. Re:And how many Solaris VM's can I run? by db · · Score: 1

      Beginning with the E10000, servers are able to run multiple "domains". I.e., the E10000 has 16 system boards when fully loaded. You can devote any combination of these boards to running different VM's. 2 boards running one solaris instance, the other 14 running another. 16 seperate instances. 4. 8. 3 and 13. You choose. This isnt quite as "virtual", but it is dynamically reconfigurable.

      I believe this carried over to all of Sun's mid to high-range servers (Serengeti, StarFire), and it's definately something they're working towards.

      I believe the difference is that it's not so much in the OS software as it's in the firmware on the machine itself.

      Dave

    2. Re:And how many Solaris VM's can I run? by norton_I · · Score: 2

      IBM is largely targeting the markets where they number of instances > number of CPUs. Or, where you want peak loads of different applications load balanced dynamically (ie, without having to say "move this processor to that node").

      As far as I know, Sun's dynamic domains are dynamic in the sense that they can be reconfigured on the fly, not that they automatically move resources where they are needed.

  19. Can't blame em... by iNiTiUM · · Score: 5, Interesting

    For a company that is planning on dropping all support for x86 in the first place, does this really surprise you? as a sun tech myself i totally see there point. Especially when the mainframes they refer to require another proprietary OS to run on top Linux. The article makes some good points, but this is also standard sun marketing.

    Sun: A solution looking for a problem

    --
    When encryption is outlawed, ou++1!@(93j++js-d9298yIUH(*Y24JKB!~
    1. Re:Can't blame em... by sxpert · · Score: 4, Informative

      ok, I have to answer this bull alltoguether

      I'm an IBM tech (not speaking for IBM, of course) and can tell all that the "proprietary Operating system" they are talking about is not really an OS, more like a virtual machine that handles all other oses. This virtual machine is n implementation of the S390 architecture that will handle several "virtual computers" at the same time by the use of multiple "virtual processors", each one used for a "virtual computer", with each having it's own PC, virtual memory handling and all.

    2. Re:Can't blame em... by yesthatguy · · Score: 1

      Can you essentially/actually/theoretically run different OSes on each of those "virtual computers"? This is, of course, assuming the OS was written and compiled to work on the architecture. That could be pretty neat, having one machine that masquerades as a bunch of different types of computers.

      --
      Yes! That guy!
    3. Re:Can't blame em... by afidel · · Score: 2

      Can you essentially/actually/theoretically run different OSes on each of those "virtual computers"

      Yes you can, in fact until the Z/800 that was the only way to run S/390 Linux. You partitioned the mainframe into multiple partitions and then loaded as many linux images on the linux partition as you liked/it could handle.

      --
      There are 4 boxes to use in the defense of liberty: soap, ballot, jury, ammo. Use in that order. Starting now.
    4. Re:Can't blame em... by Anonymous Coward · · Score: 0

      Actually, they are only dropping support for x86 Solaris - they are still selling the Cobalt qube, or whatever it's called, and also selling x86 boxes with Linux. Just read it on their web site.

    5. Re:Can't blame em... by Tony-A · · Score: 2

      Can you essentially/actually/theoretically run different OSes on each of those "virtual computers"?
      Yes. Very yes.
      Including running VM on top of VM.

    6. Re:Can't blame em... by unclefucknut · · Score: 1

      It's like VMWare, but at the hardware level. (And better I suppose)

    7. Re:Can't blame em... by Anonymous Coward · · Score: 0

      Oooh, an IBM tech and a Sun tech.

      Fight... Fight... Fight...

  20. Is this a suprise? by re-Verse · · Score: 2, Insightful

    Microsoft says the same thing. Does this make it true? No

    If anything, this is a really good sign for the ever maturing linux operating system. Of Course sun would want to move people away from an open source, free operating system, over to their 'paid for' one. And if they can't do that by simply saing "don't use linux, use solaris", it makes a lot of sense for their marketers to simply say "don't use linux, its bad... and scary". It still cuts out a potential threat to them.

    I figure if IBM says that IBM is ready for linux, i will trust that a lot more than solaris saying IMB isn't ready for linux.

    Not that i have anything agasint sun, or solaris.. i respect sun and what its doen, and been through.. i just question the reasoning for this 'article'.

  21. Sun says our product is better that other product by Penrod+Pooch · · Score: 1

    News at 11.

  22. No, it's marketing by TheReverand · · Score: 2

    There is actually a difference, apparently. You see, trolling involves making up facts to support an argument, whereas marketing involves, erm...

    1. Re:No, it's marketing by frozenray · · Score: 1

      You see, trolling involves making up facts to support an argument, whereas marketing involves, erm...

      ...two drinks minimum.

      SCNR - and kudos to Scott Adams for this one.

      --
      "There are already a million monkeys on a million typewriters, and Usenet is NOTHING like Shakespeare." - Blair Houghton
  23. It would go better on SatireWire by PHAEDRU5 · · Score: 2

    You know, like this.

    --
    668: Neighbour of the Beast
  24. That was the original idea by Anonymous Coward · · Score: 0

    running linux on x86 was the original idea of linus
    back in 1991.
    linux has come a long way from it. if linux can
    run on sparc, alpha , x86, PPC
    why cant it run and scale in a VM environment ??
    say what u may mainframes are mainframes, and
    add linux to them u have a win win situation.

    sun would be better off showing strengths of
    its own product line than try something like this.

    reminds me of the M$ FUD about Embeded Linux and
    Embede Win XP.

  25. sun un by nikkatsu · · Score: 1

    yawn. yaaaawwwwnn. wouldn't it like be funny if they said the exact opposite, hunh, wouldn't it? wouldn't it be funny?

  26. IBM milking mainframe monopoly by ChrisRijk · · Score: 5, Informative

    Just a little FYI:

    With Amdahl checking out of the mainframe business it seems IBM has decided to raise mainframe prices significantly - it's actually charged more for the same performance in 2001 than in 2000! This is why IBM's mainframe revenues increased by a fair bit between 2000 and 2001 (while it's PC and Unix revenue dropped). Mainframe revenue accounts for about half of IBM's total server revenue...

    1. Re:IBM milking mainframe monopoly by Anonymous Coward · · Score: 1, Interesting
      Guess what? IBM mainframe revenues have grown 5-fold in the same time frame, this growth at the height of a recession. In the same period we have seen many other companies die or leave the Free software business.

      You can't argue with success. It is clear as day that Sun is getting eaten up by Linux, and old rival IBM. Sun was caught with its pants down. Instead of Sun offering a comparable product, their cupboard is bare. All that is left for Sun is name calling without substance.

    2. Re:IBM milking mainframe monopoly by Mittermeyer · · Score: 2, Interesting

      Several excellent points have been made about Unix and IBM being of very different environments both in terms of workloads and skill sets. However I think you need to see the Bigger Picture, which relates to this poster's point about Amdahl, and the market pressure of the Unix servers and right behind them Microsoft.

      IBM is milking their big win over Amdahl and Hitachi, a 30-year-plus war that took a final round of reengineering costs associated with going 64-bit (keep this in mind- could be replayed with AMD and Intel). This is fine for the Three Initial Companies that can afford this, but the future of the mainframe is in doubt, because of

      *the easier entry into Unix/WiNT servers,

      *the relatively slow speed of application development with traditional mainframe tools (sometimes that's a win, you don't build the pyramids with a mobile home mindset),

      *the lack of z/OS personnel being churned out due to a mindshare failure/lack of mainframes at universities/tech schools,

      *and new applications not being created that will bring in the new business.

      So IBM needs Linux badly to grow the new businesses into the mainframes, keep the profits up with higher production runs, get those wacky penguin kids to develop new apps that sell boxes and in general keep the gravy train running even if z/OS fails due to lack of personnel/ business/ coolness whatever.

      Sun, AIX, MS and Company are already nipping at the heels of the low-end part of the MVS line, so IBM drops those prices dramatically. As Moore's Law continues it's inexorable ways, IBM will be forced to slash the low-end to medium box prices to keep the TCO in line. So many companies that might eat the conversion costs to get off the expensive hardware may stay and be inadvertent beneficiaries of the Price Wars.

      The power failure aspect of mainframes is priceless. If there is a power failure in the mainframe world all you have to do is wait until the lights are on, power the mainframe and associated SAN back on, and chug along while the Unix tribe fscks the night away. I'm sure journaling helps (gosh, had versions of that on mainframe DBs since the 80s), but nothing beats true bulletproof hardware.

      IBM mainframes use every single ounce of processor- we can top out at 100% and everything just runs slow instead of at risk of dying. Not sure whether 390 Linux will handle it but the underlying VM sure will.

      Another factor not mentioned is that mainframe software is charged by the MIPS, not site-licensed, so each move up in processor costs you far more in software costs then merely the machine and support. The Linux 390 environment doesn't have that MIPS exponential cost curve, so that aspect looks mighty tasty to sites that want mainframe bulletproof hardware and relatively cheap Linux software implementation.

      Finally, we have a person who straddles the world of UNIX and MVS (he HAS to- MVS requires Unix Services to function in a non-SNA world). Our Unix manager was a former mainframe field engineer, as was our current Network Services Director. I've had three OS environments shot out from under me, and I haven't been relegated to the scrap heap. My career goal is to be able to sysadmin in both JCL and Perl, MVS and Linux.

      Have a little flexibility guys, it will stand you in good stead over the decades.

      --
      ________________________________________ History Must Not Fall Into The Wrong Hands ___________________________________
    3. Re:IBM milking mainframe monopoly by shani · · Score: 1

      Sun was caught with its pants down. Instead of Sun offering a comparable product, their cupboard is bare.

      I've been working in the IT industry for almost 9 years now, and I can assure you that Sun has been arrogant, overpriced, and mediocre in quality the entire time I've dealt with them. I eagerly await the company's demise, and am glad that Linux will help that along.

    4. Re:IBM milking mainframe monopoly by harakh · · Score: 1
      The power failure aspect of mainframes is priceless. If there is a power failure in the mainframe world all you have to do is wait until the lights are on, power the mainframe and associated SAN back on, and chug along while the Unix tribe fscks the night away. I'm sure journaling helps (gosh, had versions of that on mainframe DBs since the 80s), but nothing beats true bulletproof hardware.

      You really think a mainframe can get a powerfailure?! Seriously - you ever been to a site that has even a smaller mainframe? most sites like that have multiple huge UPS-systems and if the system is abit more important they have backup generators aswell. The tech's taking care of the mainframe are really sloppy if they get a mainframe to go down in an unorderly fashion.

    5. Re:IBM milking mainframe monopoly by Mittermeyer · · Score: 1

      The situation was that the UPS was down and waiting on a part being flown in, so we were on city power. Needless to say Mother Nature decided to kick up her heels and lay on a storm that a) delayed the jet's arrival with the part, and b) knocked out our site. The vulnerability was about 10 hours, so it was a risk that we lost. But better the UPS was offline and we took this hit then have the UPS fry and send a killer spike down the line to toast our PDUs.

      In our industry the cost of multiple UPS would be prohibitive. This has happened twice in 20 years at the same site- the second time was lack of communication for a planned power switchover.

      Bunkie, I got some bad news- anything can cause the power to go out, and multiple UPS is not a magic cureall answer. Humans are still involved, and that extra mainframe magic mitigates even the worst decision-making going.

      --
      ________________________________________ History Must Not Fall Into The Wrong Hands ___________________________________
  27. Obvious by johnhyland · · Score: 0
    What do the folks on Slashdot think?

    Come on. What do you think the folks on Slashdot think? I don't know the first thing about mainframes, but I assure you that the folks on Slashdot think that any suggestion that Linux is not the best solution to everything is pure FUD, Linux r0x0rs, rah rah, the unstoppable might of open source will surely roll over any puny opposition unwise enough to stand in their way.

    That may or may not be true in this case - like I said, I don't know the first thing about mainframes and I haven't even read the article. I'm just saying it's pretty silly to ask slashdot if anyone should use Linux.

  28. Yea Right, what ever!! by Anonymous Coward · · Score: 0

    Sun's bull shit. Just trying to make Solaris look good. But we don't buy it for one moment, do we!

  29. CCO? by selan · · Score: 4, Insightful

    I'm not sure what to think about the fact that Sun has a "Chief Competitive Officer." Please tell me that there's more to the guy's job than spreading FUD about the competition.

    1. Re:CCO? by pnatural · · Score: 2, Funny

      Next he'll be telling us that Sun(tm) Java(tm) is a fully Object-Oriented Systems Language, Sun(tm) J2EE(tm) is a Standard, and that they, Sun(tm), do not ignore the Sun(tm) Java Community Process(tm).

    2. Re:CCO? by sean23007 · · Score: 2

      The purpose of the CCO is to blame Microsoft for having a Chief Competitive Officer, regardless of whether or not they actually have one.

      Remember, everyone is innocent but a certain Gates; thus it matters not what one does, as long as he is else.

      --

      Lack of eloquence does not denote lack of intelligence, though they often coincide.
    3. Re:CCO? by Anonymous Coward · · Score: 0

      he runs m$'s Corporate Obfuscation of Competive Kernels or COCKs for short

    4. Re:CCO? by 21mhz · · Score: 1

      It's just the exported name for "Chief FUD Officer".

      --
      My exception safety is -fno-exceptions.
    5. Re:CCO? by miniver · · Score: 2
      I'm not sure what to think about the fact that Sun has a "Chief Competitive Officer."

      What can I say? It would have been cooler to call him the "Chief Fudmeister", but he wouldn't take the position until Sun changed the job title to three words.

      --
      We call it art because we have names for the things we understand.
    6. Re:CCO? by Kirruth · · Score: 1
      It would have been cooler to call him the "Chief Fudmeister"

      It would have been, but then he might have been confused with Scott McNealy, who occupies the "Chief Suckmeister" position.

      --
      "Well, put a stake in my heart and drag me into sunlight."
  30. Misrepresented article.. by XaXXon · · Score: 5, Insightful

    This article is misrepresented as bashing Linux. It doesn't say that Linux isn't up to the job of running on a mainframe as much as it says that many of the benefits Linux offers are lost when running it on such a system -- basically bashing IBM's solution, not Linux.

    Finding mainframe staffing is an obstacle in many organizations(6); combining mainframe and Linux staffing further complicates the matter. Running multiple Linux images still requires administration that needs to grow with the number of images being run.

    This statement applies no matter what operating system you choose, you still have to find people who know the hardware. And as with all VM systems, you have to actively administrate each image. This statement is Linux agnostic.

    Although z/VM can start and stop Linux images, it cannot dynamically add resources to match demand. As a result, a mainframe would need to size for peak demand just as the Linux farm would; high utilization is a myth.

    Again.. Linux isn't repsonsible for the machine not being able to dynamically allocate resources to over-utilized images, it's a hardware/underlying OS issue.

    Applications that run on Linux for Intel need to be recompiled and recertified for each new platform; thus the application portfolio to run Linux on a mainframe is small

    Duh. It's a different architecture.

    So, SUN isn't really bashing Linux, they're bashing their competitor, IBM. No real news here. SUN is very careful not to say "Linux sucks", because they have Linux offerings, they're just saying that customers should buy the SUN/Solaris solution for their high-end systems, not the IBM/Linux solution. I'm sure we'll see something from IBM soon.

    --XaXXon

    1. Re:Misrepresented article.. by mozkill · · Score: 1

      good analysis of the article. :-) i agree with you.

      --

      -- Betting on the survival of the media industry is a serious risk. I advise investing elsewhere.
    2. Re:Misrepresented article.. by sxpert · · Score: 4, Informative

      Although z/VM can start and stop Linux images, it cannot dynamically add resources to match demand. As a result, a mainframe would need to size for peak demand just as the Linux farm would; high utilization is a myth.


      This is total bullshit, check the IBM RedBooks on z/VM

    3. Re:Misrepresented article.. by jaustin · · Score: 1

      Excellent point. I'm not sure why everything has to turn into an OS religious debate. Sun is primary interested in selling their systems and not their OS. If fact, I'm not sure why anyone would even have to pay for Solaris unless maybe you buy a Sparc clone. Their real interest is selling their Netra boxes for web services and the Sun Fire/Enterprise boxes for mainframe sized servers. As far as the OS, if it runs your stuff, who cares? Get whatever is best tuned for the hardware. It's more important to have source available at the application level than the core OS level. As long as the OS follows an open standard, your stuff is portable.

    4. Re:Misrepresented article.. by Pitchfork · · Score: 1

      I interpretated the Sun statement as follows:

      you have either 500 linux 1 HE boxes or 1 mainframe with 500 partitions. You have a total peak for all 500 systems and you must have hardware to deliver to this peak. z/VM cannot add processing power to the system as a whole, it can only shift it between the virtual machines.

      My opinion about that:

      It all depends whether the peaks of the 500 systems happen at the same time or if the peaks are distributed in time.

      When they happen at the same time, then nothing but raw processing power can help you. There is no shifting of CPU cycles between VMs possible, everything is eaten up.

      When the peaks happen at different times, the resources from one VM can be shifted to another VM and IBM has a solution to minimize the costs.

      The solutions Sun offers is also viable: take 500 cobalts put a load balancer in front and let all the 500 cobalts run ALL applications/web sites. In this way the processing power is also perfectly shared between the instances. With the negative impact that they is no real barrier between the systems and that will make some of your customers unhappy.

    5. Re:Misrepresented article.. by sjames · · Score: 2

      When the peaks happen at different times, the resources from one VM can be shifted to another VM and IBM has a solution to minimize the costs.

      Absolutly! And if you're big enough to need a mainframe, you probably have tons of batch jobs that can be queued at low priority to take advantage of every little dip in demand.

    6. Re:Misrepresented article.. by PerlPo8 · · Score: 1
      I agree that this article is being blown out of proportion. I think Sun is merely positioning themselves so their sales staff has ammo to promote Sun solutions. This FUD could be used to promote Sun Hardware running Solaris or Linux--"Our hardware can run Linux natively, not on top of a VM".

      I'd just like to add that, having worked for several years in a strongly IBM Midrange-oriented shop, my observation of IBM's technical goals is that they seem to put a greater emphasis on stability than on performance. I suspect that their Linux solution is quite solid, but a little slow. That about describes every IBM Solution I've ever used.

      --

      --
      "I'm don't know exactly what an AS/400 is, but I'm pretty certain I wouldn't want one up my ass" --Lou

    7. Re:Misrepresented article.. by YetAnotherDave · · Score: 1

      One quote stands out:
      'This is Linux. It's designed for Intel. '

      I think that the kernel developers who work so hard to make linux work on so many different platforms could reasonably be a bit bothered by this statement...

    8. Re:Misrepresented article.. by Sid_Boyce · · Score: 1

      I don't know what Sun really thinks, they are all over the place on their thoughts. Mr. McNeally never misses an opportunity to knock the Mainframe as obsolescent technology, yet Sun's current technology takes you back to the mainframes of the early 1980's. I also hear that mainframes are soooooo expensive, yet we look envious at the cost and maintenance prices paid for Sun equipment that largely act as front-ends to the mainframe workhorse. You don't get those sorts of prices for mainframes.
      My guess is that Sun is seeing unaccustomed competition coming from equipment that's decades ahead in design, performance, resiliance, maintainability and reliability.
      Ever noticed there is never a reasoned argument, we see throw-away comments with flippant use of words like "obsolete", "dying", "old", etc. in sentenses that read more like phrases.
      Regards

    9. Re:Misrepresented article.. by Maddog_Delphi97 · · Score: 1

      And it's not an entirely inaccurate statement; the initial development of the Linux WAS on an Intel system (An early model of the IBM PS/2 system, if I remember correctly).. most of the cutting-edge development of the Linux is still geared specifically Intel..

      And the truth, having Linux run on non-Intel systems have never really been the first priority.. some of the best development of non-Intel Linux are done by non-profit organizations such as Debian (Alpha development and Motorola developements are two good examples of this).. meanwhile, companies like Red Hat have recently de-emphasized keeping non-Intel offering on parity with Intel offering mostly because it not profitable enough to otherwise do so.

      And dont forget, the kernel source code for Yellow Dog Linux (a distro of Linux geared toward MacIntosh) is VERY different than the kernel maintained by Linux Torvalds. I think the main reason for this is that Linux doesn't seem that much interested in doing testing and debugging on MacIntosh hardware, so Mac-only Linux development is always a step behind Intel development.

      On an almost unrelated note, I've heard rumors that Linux runs faster on Sparc hardware than Solaris.. is there any truth to this? I haven't seen any hard benchmark numbers to confirm or deny this..

    10. Re:Misrepresented article.. by ibeleo · · Score: 1

      If you read the few sentences before this one, you will see Shahin Khan (Sun author) was discounting the idea you can take many low utilization boxes and consolidate them onto the mainframe, end up with higher utilization and still expect to handle peak loads. The many under utilized boxes have the spare capacity in case there is a surge in traffic. That's what he means by sizing for peak demand.

      He could have made the point clearer, I don't think he was denying z/VMs ability to shift resources between Linux images.

    11. Re:Misrepresented article.. by BlowCat · · Score: 1
      Yes, Linux runs faster that Solaris, at least on low end machines. My Sparc 3 was almost useless (i.e. terribly slow) until I installed Linux on it.

      I believe that Solaris is faster on SMP systems, but I have never had a chance to compare.

    12. Re:Misrepresented article.. by tigga · · Score: 1

      One of problems is a GNU C which really sucks on platforms other than Intel...

    13. Re:Misrepresented article.. by Anonymous Coward · · Score: 0

      Not even - solaris SMP is more _scaleable_ than linux SMP, but that doesn't mean any single task will run faster, just that on a box with lots of processors and lots of tasks, solaris won't waste as much time deciding what should run where...

    14. Re:Misrepresented article.. by Anonymous Coward · · Score: 0

      And that's exactly what misses the point (this being marketing, obviously Khan knows this and is hoping that you won't).

      You can have S servers with x% overcapacity, or 1 mainframe with S VMs with y% overcapacity.

      Unless you ever have situations where every single of those S servers has a full load spike at once, S*x > y and the mainframe wins.

      Having all S servers max out at once is incredibly unlikely, and in the mainframe case will results in temporary slower performance until some or all of the spikes subside.

      Even if you are always running at full capacity (the degenerate case), S*x == y, and you still win because you have less machines to manage.

      Khan obviously can't make the point clearer, because he's relying on misleading people into making incorrect conclusions without explicitly (well, provably anyway) lying. Its called marketing, and all companies do it.

  31. You know when it's a good idea ... by linuxdoctor · · Score: 1

    ... when you have detractors.

    Well, they said that man could never fly and that he would never make it to the moon. Linux on mainframes IS a good idea, especially when someone says that it isn't.

    So what if those who have a vested interest cook up reasons that say it can't or shouldn't be done? The more they rail against it, the more someone will find a good reason that it could and that it should.

    Let them fume and sputter all they want. Linux has already changed the world, and for the better too. Sun's self interested objections leave me unmoved.

  32. How do you get this job? by JordanH · · Score: 1
    • ...Shahin Khan, Sun's chief competitive officer...

    Is there an MBA with a concentration in Competition?

    Do you have to be veteran of an Olympic team?

    Is a qualification for the chief competitive officer that he/she be loud and has to interrupt a lot?

    Doing a quick Google Search, I see that Palm has one of these. The Sun guy didn't turn up there. Must be a new thing.

    1. Re:How do you get this job? by Anonymous Coward · · Score: 0

      Hey, hey. Shahin Khan is a Sand Jemimah. In other words a filthy AY-rab.

  33. Linux isn't the target of the bashing? by jone1941 · · Score: 1
    This story isn't about bashing linux and it's place on a mainframe. It's on the use of a propriotary os to provide a virtual machine for any os not designed to be run in a virtual maching in the first place.

    No need to start flame wars about how sun is trying to knock Linux (mostly because there isn't anything to knock).

    It should be noted though, the fact that they compare pricing of a 1 cpu mainframe to a 1U rackmount is insane. What they don't talk about is the fact that
    • NO ONE
    buys a single cpu room sized mainframe (at least now that venture capitol has dried up).
    --
    Fear trumps hope and ignorance trumps both
    1. Re:Linux isn't the target of the bashing? by erasmus_ · · Score: 1

      I'm pretty sure the article mentions the _per-processor_ cost, rather than comparing single-processor systems only. Which of course is still incredibly high on a mainframe.

      --
      Please subscribe to see the more insightful version of th
    2. Re:Linux isn't the target of the bashing? by Wdomburg · · Score: 2

      >buys a single cpu room sized mainframe

      Not only that, but mainframes don't take up a whole room anymore.

      The Z/Series starts at a footprint of about 14 square feet and goes up to 30 in a two frame configuration, the S/390 is about 10 square feet, and a 42U rack cabinet is about 7 square feet.

      So even if you take the worst case scenario, the "room-sized" perception suddenly shrinks to about the size of four standard racks.

      Matt

  34. Benefits of the mainframe by kick_in_the_eye · · Score: 1, Informative

    The benefits of the mainframe are plenty. Awsome uptime, its in years, not months. Amazing I/O and storage capabilities. ESCON is unbelievable in the way it works, sharing I/O through timeslicing it. Time slicing CPUs. What you can do on the new Sunfires (Ex800 and E15K) are pretty minimal compared to the slicing and dicing of mainframe. Its more a story of hardware than software. The best solution would be Solaris on the mainframe :-)

    1. Re:Benefits of the mainframe by khuber · · Score: 2
      >The benefits of the mainframe are plenty.

      A 1000 processor Cray T3E 1200 would smoke your pathetic little mainframe :). And you wouldn't have to use crap like JCL. However, there's this little problem called cost of ownership. If supercomputers and mainframes were cost-effective there wouldn't be the huge push for commodity clusters, blade servers, and so on. There are probably applications where mainframes are cost-effective, but there are problems like almost nobody coming out of college with mainframe knowledge.

      -Kevin

    2. Re:Benefits of the mainframe by Blrfl · · Score: 1
      khuber writes:

      A 1000 processor Cray T3E 1200 would smoke your pathetic little mainframe :)

      In some applications, yes, I couldn't agree with you more. Mainframes suck eggs when it comes to number crunching, but even a small mainframe will blow the doors off any supercomputer when it comes to I/O intensive applications such as databases.

    3. Re:Benefits of the mainframe by RetroGeek · · Score: 1

      And you wouldn't have to use crap like JCL

      Um, its REXX now....

      --

      - - - - - - - - - - -
      I am a programmer. I am paid to produce syntax not grammar. Deal with it.
    4. Re:Benefits of the mainframe by biobogonics · · Score: 1

      The benefits of the mainframe are plenty. Awsome uptime, its in years, not months. Amazing I/O and storage capabilities.

      Steve Comstock, in comp.lang.asm370, posted a link to a document he wrote from the IBM point of view which discusses how Unix is hosted on z/os. It also does a good job of describing MVS and z/os for those from the Unix side. Considering the architecture that Unix and Linux were written for, it's a wonder that it *can* be run in a mainframe environment, given how different the mainframe really is. For example, a mainframe really doesn't have the same type of terminal session as a Unix system and batch jobs are really alien to Unix.

      see http://www.trainersfriend.com/Papers/zos_unix.pdf

    5. Re:Benefits of the mainframe by LinuxHam · · Score: 2

      but there are problems like almost nobody coming out of college with mainframe knowledge.

      Which is why IBM is trying to save the mainframe by running Linux on it. Suppose a company buys a z box running Linux. They get one 48U rack for the system and another for the DLT tape libraries. They contract out one mainframe support guy from IBM, and hire 8 cheap CS college grads to run the Linux virtual servers. Or one seasoned vet and 6 kids. All IBM has to do is guarantee job security for a few new hires a year by giving them mainframe training.

      --
      Intelligent Life on Earth
    6. Re:Benefits of the mainframe by rogueroo · · Score: 1

      JCL continues to be used as the only acceptable way of submitting programs to JES internal readers. Even UNIX System Services under OS/390 and z/OS contains daemons that are controlled and monitored by JCL procedures (i.e started tasks) executing within address spaces created by submitting JCL to an internal reader!

    7. Re:Benefits of the mainframe by cyber-vandal · · Score: 2

      He's referring to the common clueless assertion that mainframes are crap by comparing JCL (a set of control statements) to awk (a scripting language). JCL does suck, but it's very well documented, unlike too many Unix functions (and no, confusing man pages don't count).

    8. Re:Benefits of the mainframe by cyber-vandal · · Score: 2

      almost nobody coming out of college with mainframe knowledge


      And of course you would trust your Solaris cluster to someone who'd only used a Sun workstation at college/university. Or have you never heard of training?


    9. Re:Benefits of the mainframe by khuber · · Score: 2
      >And of course you would trust your Solaris cluster to someone who'd only used a Sun workstation at college/university. Or have you never heard of training?

      Nice straw man! If that were anything close to what I said you would've totally kicked my ass.

      I'm just listing one disadvantage of using mainframes at a company -- that colleges don't typically use them or teach about them. I never said only college grads administer computers or that nobody needed training after college.

      -Kevin

  35. What about AIX? by YourMissionForToday · · Score: 0
    I've worked with Solaris, AIX, and Linux, and AIX is my favorite OS of the three (my company's inventory database runs on AIX). I love SMIT, it's lean, text-based, yet very user friendly.

    I admit that I don't have any experience running Linux on Big Iron, but it seems that these custom, proprietary Unices always work great on their hardware.

    Linux was originally intended to be a cheap way to get that UNIX-type functionality (running it on commodity hardware). Putting it on big iron seems backwards to me.

    1. Re:What about AIX? by Cheetahfeathers · · Score: 1

      smit is definitely a kick ass config tool. I wish more OSes had a similar config tool. The text mode of smit is very good. I especially love that it (optionally) shows you the command line versions of what it's doing behind the scenes.

  36. FUDnews.. by josepha48 · · Score: 3, Insightful
    This sounds like FUD news.. actually it sounds like something Microsoft would spread. I guess Sun see Linux as competition. The question then becomes which is really more powerful on a cost basis: Linux/MF or Sun/Solaris. Now one would have to look at the actual cost of the MF running Linux and how it performs vs a Sun comparaible hardware running Solaris 8-9 or whatever. Yes Sun does have MF sized servers.

    Next thing to do would be to ask someone that recently switched to linux on the mainframe, like ebay... hope one of the links below still works...

    http://www.cio.com/archive/010101_et_content.htm l

    http://abcnews.go.com/sections/tech/DailyNews/ib ml inux000517.html

    http://www.zdnetindia.com/biztech/resources/ebusin ess/ecommerce/stories/45234.html

    --

    Only 'flamers' flame!

  37. Makes sense to me by NitsujTPU · · Score: 2, Insightful

    People these days seem to forget about the overhead of interpretters and virtual machines. If the article is correct, then the z800's running zVM emulate Intel x86 architecture in order to run Linux. Heck, even poorly written native compiled code generally has advantages over such a set up.

    There are however, notable exceptions, given the nature of mainframe processors, if all of your apps are written unoptimized for such a system, then you would want to unify them in a familiar abstraction, given a close enough match, this makes Linux a natural choice. Of course, why would you buy an expensive mainframe and not optimize for it?

    To the naysayers slamming Sun as merely trying to boost SunOS, well, yeah, they are, but lets look at the situation.

    1) Sun still has SunFire servers, which are QUITE powerful.
    2) Solaris is no longer competing with HP-UX, since HP-UX is no more. Sun sells windows and linux based solutions. In other words, Sun has no reason to just blindly nay-say against Linux. As far as exploiting Linux for being a hot technology, well, they're doing that too. That's business for you, you gotta do what you gotta do.

    In otherwords, the z800 isn't exactly slaughtering Sun's business, but you gotta have whitepapers to back up your statements when you're bidding to large customers. Saying "just cuz" isn't good enough. Sun's scoring one for the people who want to buy their products. It's not "slamming linux."

    1. Re:Makes sense to me by Steveftoth · · Score: 2, Informative

      Do you even know what kind of VM IBM is talking about when they talk about their zSeries?
      I'll use tech that you are probably familar with. It's exactly like VMWARE for windows/linux. Except that the zSeries OS that runs the virtual machines runs at a much lower level then the VMWARE program and the HW is optimized for the execution of VMs. Unlike the x86 on x86 emulator that VMWARE does, the zSeries boxes run their VM's very fast. All the code that is executed runs native to the processors in the box, you have to recompile everything to run on them.
      see this url for more facts on IBM's stuff. I'm not saying it's the answer, but this paper from SUN is FUD.
      http://www-1.ibm.com/servers/eserver/zseries /os/li nux/facts.html

    2. Re:Makes sense to me by mjstrom · · Score: 1

      People these days seem to forget about the overhead of interpretters and virtual machines. If the article is correct, then the z800's running zVM emulate Intel x86 architecture in order to run Linux. Heck, even poorly written native compiled code generally has advantages over such a set up.

      I don't think that's the type of virtual machince the article talks about. The VM in this sense is like VM Ware were it controls access to the actual hardware and gives the OS a "virtual" machine to run on. However, it doesn't "emulate" x86 architecture to run linux, linux runs natively on the hardware of the mainframe. It iss not a Java VM.

      1) Sun still has SunFire servers, which are QUITE powerful.

      Uhm, yes, they are powerful. They can juggle more things at once and keep plugging. BUT, and this is the important part, on individual processes (apache) that don't multi-thread (a lot of commerical apps) they are SLOW.

    3. Re:Makes sense to me by chewbca · · Score: 1

      2) Solaris is no longer competing with HP-UX, since HP-UX is no more.

      umm.... think you might be confused between PA-RISC and HP-UX. PA-RISC is being phased out in favor of IA-64. But HP-UX is alive and well, thank you very much :P, and being ported to IA-64. HP isn't competing with Sun anymore because IBM is the logical target now, with their services business.

      --
      -- "This is my sig... there are many like it but this one is mine"
    4. Re:Makes sense to me by NitsujTPU · · Score: 2

      Really? Cool. I heard that HP cut their entire staff for development. Cool...

    5. Re:Makes sense to me by zm91019 · · Score: 1

      People these days seem to forget about the overhead of interpretters and virtual machines. If the article is correct, then the z800's running zVM emulate Intel x86 architecture in order to run Linux.

      Interpretation not correct -- article playing upon ambiguity. zVM does not emulate x86. 390/Linux is a real port to the zVM virtual machine. Optionally, you can run Linux natively on the bare S/390 hardware or in an LPAR or under the Virtual Image Facility. Virtualization overhead is not a consideration here.

    6. Re:Makes sense to me by pmz · · Score: 2

      Uhm, yes, they are powerful. They can juggle more things at once and keep plugging. BUT, and this is the important part, on individual processes (apache) that don't multi-thread (a lot of commerical apps) they are SLOW.


      Sun servers are not "SLOW" for the reason you give. Apache will still scale with the number of processors in a Sun server, regardless of whether each process is threaded. Also, Sun UltraSPARC processors ship with large caches that can hold most of an Apache server process, and the SPARC architecture is pretty much designed for C-language-programmed UNIX kernels and applications, such as Apache.

      So, they are not "SLOW" as you claim but are actually quite FAST. For example, if you have a server with enough processors to delegate one Apache process per processor, the kernel can keep every processor busy with minimal context switching or threading overhead. And recent Solaris kernels are extremely good at keeping all processors busy very efficiently. Under really really high load, Solaris on SPARC is hard to beat.

    7. Re:Makes sense to me by Spy+Hunter · · Score: 2
      If the article is correct, then the z800's running zVM emulate Intel x86 architecture in order to run Linux.

      No, you misunderstood. The VM doesn't emulate an x86, it emulates lots of smaller mainframes on one big mainframe. Like a beowulf cluster in one box. Emulating an x86 would be silly and slow.

      --
      main(c,r){for(r=32;r;) printf(++c>31?c=!r--,"\n":c<r?" ":~c&r?" `":" #");}
    8. Re:Makes sense to me by chewbca · · Score: 1

      yes you heard about the PA-RISC design team, they were let go (actually they all went to Intel, ostensibly to help design IA-64 chips, I guess) but HP-UX dev is humming right along :))

      --
      -- "This is my sig... there are many like it but this one is mine"
    9. Re:Makes sense to me by Doomdark · · Score: 2
      Sun sells windows and linux based solutions

      Minor nitpicking; AFAIK Sun does NOT sell Windows-based solutions. The only thing related to Windows Sun sells are certain software packages (Forte, StarOffice) that have Windows-version available along with Solaris and Linux-versions. Sun's products can certainly be integrated with Windows systems etc, but Sun doesn't sell such systems.

      --
      I like paying taxes. With them I buy civilization -- Oliver Wendell Holmes
    10. Re:Makes sense to me by NitsujTPU · · Score: 2

      I saw some Sun boxen being demoed running NT. I was fairly sure that this was an effort on the part of Sun Microsystems.

    11. Re:Makes sense to me by sjames · · Score: 2

      People these days seem to forget about the overhead of interpretters and virtual machines. If the article is correct, then the z800's running zVM emulate Intel x86 architecture in order to run Linux. Heck, even poorly written native compiled code generally has advantages over such a set up.

      A VM is not really an emulator at all. The idea is that the VM virtualizes the the system through special exception handling, not unlike the way any multitasking OS virtualizes memory, abstracts hardware through drivers and transparently switches contexts. IBM hardware is designed with this in mind, so doesn't take as much of a hit as x86 would, and doesn't have to resort to full emulation.

      It's more comperable to Win4Lin or Mac On Linux than to an emulator.

      So, it's actually running Linux compiled native for s390 arch.

    12. Re:Makes sense to me by foonf · · Score: 2
      I saw some Sun boxen being demoed running NT. I was fairly sure that this was an effort on the part of Sun Microsystems.


      You're sure it wasn't a Sparc box with one of the add-on PC compatibility cards (which are really seperate computers on a card, with an Intel CPU, dedicated RAM, etc.)?
      --

      "(Man) tries to live his own life as if he were telling a story. But you have to choose: live or tell." --Sartre
    13. Re:Makes sense to me by mjstrom · · Score: 1

      True, solaris scales well. One box can handle more of a load than an equivalent intel box (or at least til the sun runs out of memory and you die a painfull death of swap-storm). But, by speed I meant on a process level - the same request will finish faster on intel all else being equal - the sparc hardware just isn't as quick as intel in terms or processing speed. But of course that depends on what you are running on the box, its memory usage, how it uses resources, etc. That and sparc hardware isn't cheap, esp for what you get.

      All in all, I think I would rather have linux on a mainframe then a high end sun machine.

    14. Re:Makes sense to me by NitsujTPU · · Score: 1

      I didn't really spend that much time on the NT demo. It could be.

    15. Re:Makes sense to me by NitsujTPU · · Score: 1

      Right, I commented on this earlier. I thought that they meant VM in the JVM sense, not as in multiplexing hardware.

    16. Re:Makes sense to me by Anonymous Coward · · Score: 0

      Ummm...no.

      z/VM isn't about tricking Linux/x86 into thinking that it's running on a PC. It's about tricking 32-bit LinuxPPC into thinking that it's running all by itself on a PPC (and it is running on a PPC instruction set).

      It's a moot point anyway. In 2Q2002, 64-bit Linux will be released by SuSE and RedHat for zSeries, pSeries, and iSeries.

  38. Hands-on experience with Linux on a mainframe by ciurana · · Score: 5, Informative

    I agree mostly with the article because I recently evaluated an IBM mainframe against an AIX SP2 and a Solaris 4-processor server. Most of the issues in the article, particularly performance, are right on.

    The application we were testing was extremely processor and memory intensive. While there was a web component, the biggest problem was moving a large number of bitmaps in one format into the server, convert them from base 64 to a binary representation, rasterize them, and convert them to a "browser friendly" format such as JPEG, GIF, or PNG. We had to complete hundreds (> 200) of these operations per second.

    I really wanted to use Linux because most of my staff is familiar with it and our customer felt warm and fuzzy about using IBM equipment. At the end of the day, however, the Linux mainframe only gave us 25% of the minimum speed that we needed for our process to be successful. IBM and a certain German Linux company tweaked everything they could but the performance wasn't there. The AIX vs. Solaris match was more evenly paired. My customer decided on Solaris because they offered a few advantages in Java tools that AIX didn't have. All vendor's boxes had equivalent processor and memory configurations.

    I would like to spread the Linux credo as far and wide as possible. What we must understand is that, in order to make Linux a viable option in mission-critical applications (the kind of thing sitting on a mainframe), the performance and "hardening" of something like MVS must be present. Linux just isn't there yet.

    Disclaimer: I'm under NDA so that's why some aspects of this posting are a bit vague. Drop me an email if you want more details regarding our experience but our conversation will be "off the record."

    Have a nice wknd,

    E
    --
    http://eugeneciurana.com | http://ciurana.eu
    1. Re:Hands-on experience with Linux on a mainframe by crush · · Score: 3, Insightful

      You are talking about a different deployment than the one that is being attacked in the Sun article. What the latter is discussing are multiple images of Linux being hosted on top of a VM.

      There is no reason why you should have been doing that in your case: you should have dispensed with the VM layer and just used Linux "native".

      Basically the article is Sun bashing (perhaps righlty or wrongly, I don't know) the concept of "server farm in a box", which is completely different from your task!

    2. Re:Hands-on experience with Linux on a mainframe by Anonymous Coward · · Score: 0

      I guess ImageMagik can't do that huh.

    3. Re:Hands-on experience with Linux on a mainframe by Florian+Weimer · · Score: 2

      Isn't Sun offering "server fram in a box" (aka partitioning) for their high-end systems as well?

    4. Re:Hands-on experience with Linux on a mainframe by TheAwfulTruth · · Score: 2

      ImageMagik is ok for your single user type app or small amounts of processing. It is not at all an efficiant image processing library at all. It especually does a natsty job on memory when working with large files. It's grate and all for your basic app (We've used it rather successfully), but it has some severe limits on speed and memory use. I wouldn't use it for what that guy was talking about.

      --
      Contrary to popular belief, coding is not all free blow-jobs and beer. Those things cost MONEY!
    5. Re:Hands-on experience with Linux on a mainframe by ciurana · · Score: 2

      crush wrote:

      What the latter is discussing are multiple images of Linux being hosted on top of a VM.

      We tested running Linux in native mode and on top of VM on the z90. It was s-l-o-w anyway. By extension, running it in a VM partition would've made it slower.

      My point is not to bash Linux; I like using it (and use it exclusively for 99% of my work, from notebooks to servers), but Linux just isn't ready for mission-critical applications running on mainframes. Sun is a lot closer to that goal.

      For mission-critical stuff on mainframes, the kind of apps where you put your career on the line, MVS is still the way to go. Being a Linux chauvinist doesn't improve its performance or reliability on z90 hardware. I don't doubt it will get there; it just hasn't quite arrived yet.

      Cheers,

      E
      --
      http://eugeneciurana.com | http://ciurana.eu
    6. Re:Hands-on experience with Linux on a mainframe by crush · · Score: 2

      Yeah, but I think the difference is that IBM is suggesting co-ordinating a whole load of different images of Linux using their z/VM, whereas Sun is talking about using some form of SMP.

    7. Re:Hands-on experience with Linux on a mainframe by PoiBoy · · Score: 1

      It sounds to me that what you wanted to do was a rather compute-intensive application. My understanding, though, was that IBM's main push for Linux/zSeries is to replace file/print/web servers which typically are not so compute-intensive. Perhaps your experience was more along the lines of the wrong tool for the job in the first place.

      --
      Sig (appended to the end of comments you post, 120 chars)
    8. Re:Hands-on experience with Linux on a mainframe by crush · · Score: 2

      OK, I get your point. I wasn't accusing you of doing any bashing mate! I just thought that you were saying that you had tested in the exact situation that Khan is describing: VM hosting as opposed to LPARs. I've used LPAR'ed Red Hat on S/390 for FP number crunching and found it to be great. But, I didn't do extensive testing or comparisons with Enterprise boxes.

    9. Re:Hands-on experience with Linux on a mainframe by Anonymous Coward · · Score: 1, Insightful

      You can't really compare a zseries on computation power. This machine is extremely competitive on I/O (serving static pages, doing database queries, read/writes, etc.), but is not very strong on computation. If you want to do computations with IBM equipment, use their pSeries line, which is risc based, or their intel line - the xSeries.

      For your specific application, you - and your vendors - should have spared themselves the trouble. All the tweaking in the world would not make a mainframe with 20 CPU's that costs some $2 millions comparable to a cluster of intel boards of a similar cost. That's is why these machines are not used for modeling and rendering, but for payroll, inventory, etc.

    10. Re:Hands-on experience with Linux on a mainframe by tkrotchko · · Score: 2
      Contrary to what most people believe, you don't buy an IBM mainframe for CPU performance, rather, you buy it because of the I/O performance.


      That's what makes them expensive (well, that and the reliability).

      --
      You were mistaken. Which is odd, since memory shouldn't be a problem for you
    11. Re:Hands-on experience with Linux on a mainframe by Anonymous Coward · · Score: 0


      ImageMagick is so slow, it must've been some
      college freshman's CS project...

    12. Re:Hands-on experience with Linux on a mainframe by rfsayre · · Score: 1

      the poster didn't mention ImageMagik... perhaps he was using Java APIs, or something else enitrely. Almost anything would be better, and, judging from the poster's comment, he probably knows that.

    13. Re:Hands-on experience with Linux on a mainframe by fferreres · · Score: 1

      Works for me. I resize car press photos on the fly for various sizes depending on position and cobranded site. Of course, the originals live in MySQL and the resizes are all catched.

      --
      unfinished: (adj.)
    14. Re:Hands-on experience with Linux on a mainframe by Anonymous Coward · · Score: 0

      Its true that the IBM mainframe isn't for CPU intensive applications. There are still problems with IBM's support of their hardware under Linux. All the drivers are also closed source, so when a bug is found you have to wait a few weeks for them to make the fix public before you can allow your system to work properly. The Gigabit Ethernet system has had significant issues with linux compatability (they had to release a new driver if I remember correctly because the kernel developers wouldn't let them add in a patch so their old drivers would work) The new drivers they released weren't compatabile with the old ones, which required some re-working of every linux box using the cards. On top of this the kernel version is only at 2.4.7, so I think the local root exploit bug is still there by default ( its been a while since I cheked ). This has a chance to be interesting in the future, but I think that people should give this technology some time to prove itself. Even though IBM would like you to believe VM doesn't crash, from experience I can tell you it can (and does sometimes) and having to fix 80 linux machines cause vm crashed is not a lot of fun( expecially when you're trying to do it over 3270) There are some neat things that you can do with this system, but even web servers eat cpu, and when you have to share 9 cpus between X machines, you can definately have one machine going crazy impacting the groups performance. You can do things to limit the impact, but it is still there. One of the things that is nice is that you can dynamically allocate some hardware to linux instances, and be able to use them, but of course to make major modifications, you have to edit a flat file in the system and reboot VM. I know in my organization bringing down an entire pool of servers isn't a good idea, but if you start doing server consolidation on one of these, make sure you have a backup because you will need to shut it down to do something to it at some point. We'll see where it goes, but it has definately been a learning experience.

    15. Re:Hands-on experience with Linux on a mainframe by Anonymous Coward · · Score: 0

      I'm wondering why IBM is not releasing SPECINT, SPECFP, SPECWEB or any comparable results for their mainframes. They used to say these are not relevant on this platform, but I cannot believe why somebody would pay a couple of $100000 and not care about the performance of their apps. Your comment above seems to confirm this suspicion.

      Could you run a portable benchmark like lmbench on that system and post the results here? It sure would be interesting.

  39. The FUD heard round the world... by Duderstadt · · Score: 4, Interesting
    As the infamous Halloween Documents stated, Linux is primarily a threat to proprietary *NIX setups.


    Now, Sun offers up the ultimate proof: Linux is just fine as long as it impacts the x86 world - but don't dare put it on a platform that affects us.


    To be fair, IBM's offering is not perfect - yet. What Sun is preparing for is a future Linux and Big Iron combo that will be. They are afraid, and this FUD is the proof.

    1. Re:The FUD heard round the world... by aminorex · · Score: 1

      Well it's not FUD if it's true. I would prefer to
      address the substance of the arguments rather than
      relying on demagoguery.

      Peronsally, I think the decline, if not death,
      of Solaris is inevitable -- indeed, manifest
      already -- so that Sun will be doing the same
      think IBM is doing now, very soon, if they can
      get their egos under control long enough to
      earn their shareholders some money.

      --
      -I like my women like I like my tea: green-
    2. Re:The FUD heard round the world... by Duderstadt · · Score: 1
      Companies that are in trouble often offer up FUD as a first line of defense, and Sun is or will soon be, getting murdered by competitors.

      Competition in the low-end and nid range markets (Microsoft, .Net and Free *NIX):
      The threat here is somewhat severe. Why buy a Sun solution if you can pick up an Intel/AMD (IA-32 or IA-64) rig for half the price and run Windows on it? Or better yet, run Linux on it? Where is Sun's advantage here? Hint: it's not Java...

      Because .Net will likely kill Java. The .Net Initiative can (or so MS thinks) deliver something that Java cannot: a platform for in-the cloud-data access.

      Competition in the high-end market (Free *NIX):
      IBM, SGI, etc are offering solutions powered by Linux. Unless you just have to have the Sun logo on your boxen, Linux has been proven more than capable on relatively big iron.

      So, yeah... I think it's FUD.

    3. Re:The FUD heard round the world... by ezzewezza · · Score: 1

      so, the proof is in the ... err ... FUDding?

    4. Re:The FUD heard round the world... by Syberghost · · Score: 2

      Now, Sun offers up the ultimate proof: Linux is just fine as long as it impacts the x86 world - but don't dare put it on a platform that affects us.

      Yeah, don't dare put it on, say, a Cobalt Raq...

  40. Sounds like the same thing I'd say if... by neilb78 · · Score: 1

    I were in his shoes....

    --
    © 2004 The SCO Group, Inc. All Rights Reserved.
  41. this isnt about linux its about IBMs success by mAIsE · · Score: 2, Insightful

    sounds like they have more of a problem with IBM than with linux.

  42. Pure crap by swagr · · Score: 3, Insightful
    And Linux isn't designed to run in a virtual machine; implementation decisions that make sense on PC hardware don't fit well in a virtual machine(4). This is Linux. It's designed for Intel. It's not tuned for the mainframe hardware in which it's running.
    • Contradiction: If it's running on the VM, why should it be tuned for the hardware??? Shouldn't the VM worry about hardware tuning?
    • Does he know that the low level stuff in linux/arch was written at IBM, not some open source hacker?
    • Does he know that Linus' point in making the kernel more modular is so that it's NOT designed for intel???

    --

    -... --- .-. . -.. ..--..
    1. Re:Pure crap by Anonymous Coward · · Score: 0

      Does he know that the low level stuff in linux/arch was written at IBM, not some open source hacker?
      Well, don't forget that when IBM start hacking at a GPL'd operating system, they are essentially becomming open-source hackers ;-)

  43. What about validity of their aguments. by Petrus · · Score: 1

    All seem to be rather moot points.
    It's free layer on proprietery z/VM, no problem.
    Linux is not as optimized as z/OS, well, it has other values.
    VM and Linux engineers are different sorts - well, anynone can learn.

    But what about this one.
    Would it be true that disk cashing would be
    counterproductive?
    What about the cost issue: what is cheaper, faster, smaller, less power hungry and services more users: server farm with 100 servers or one mainframe?

    They are not really saying much against IBM,

  44. What is it with these Sun guys? by lhand · · Score: 2, Funny

    They keep shooting themselves in the foot wrt the Open Source crowd. Now the've reloaded and started shooting again. You'd think they would have run out of ammo by now.

    One of the beauties of Linux is that it can be ported to so many different platforms easily. Sun uses it and then goes on to say IBM shouldn't? wtf? There are valid reasons to run Linux in multiple virtual machines. I even do it here on my PC.

    Note to self: Must drink less coffee....

    1. Re:What is it with these Sun guys? by Arandir · · Score: 1

      Yeah, god forbid anyone should ever say anything negative about Linux, even if it happens to be true. My OS, right or wrong, everything else is Evil.

      --
      A Government Is a Body of People, Usually Notably Ungoverned
    2. Re:What is it with these Sun guys? by Brandon+Hume · · Score: 1

      One of the beauties of Linux is that it can be ported to so many different platforms easily.

      This really has nothing to do with what is being discussed here, but I thought I'd point out that Sun has repeatedly demonstrated that they thing even the PORTING step for software should be rendered moot. (Java, binary compatibility straight up the SPARC architecture...)

      Sun uses it and then goes on to say IBM shouldn't?

      You need to re-read the article. Sun is employing Linux on its Cobalt servers and the like. Sun's article is saying that it seems pretty foolish to spend a deranged amount of money on das ubermaschine and then stick Linux on it. (Scott McNealy has said this same thing even in relation to Sun's own hardware!)

      They even say it explicitly in the article... if IBM claims to get 20 separate images on a $400000 machine, WHY wouldn't you just buy twenty PCs and run Linux on them? You see this exact argument again and again on Slashdot whenever someone suggests buying Sun gear... why does it suddenly become "FUD" when Sun uses it?

      There are valid reasons to run Linux in multiple virtual machines. I even do it here on my PC.

      Your PC is not a mainframe. Beyond yourself, nobody gives a damn if it starts crashing, performance goes into the toilet, programs start malfunctioning, or what have you. Part of Sun's point is that big gear like is being discussed DOES get deployed where many people WILL care.

      --
      Brandon Hume
      hume -> BOFH.Halifax.NS.Ca, http://WWW.BOFH.Halifax.NS.Ca/
  45. Like You Didn't know the answer by Anonymous Coward · · Score: 1

    "VM/Linux combo isn't a very good match. What do the folks on Slashdot think?"

    GEE I WONDER???? It's like walking into a church and telling all the people "My friend said Jesus doesn't love you. Do you all agree with that statement?" PLEEEEEZE!

  46. Linux not ready for Mainframes by rlangis · · Score: 2, Informative

    Wasn't there a story not too long ago that mentioned how Sun was going to support Linux on lower-end machines, but NOT on the high-end Enterprise systems? (bah, I can't find the link) Anyway, people were saying "Well, Linux isn't ready for Enterprise-type systems yet, so keeping the proprietary *nices on these systems isn't a big deal."

    Now, Sun comes right out and says this, and people start complaining? Sure, perhaps Sun is trolling for /. Yeah, right.

    You may think I'm biased: I work for Sun, after all. Don't get me wrong - I'd absolutely *love* to take one of the *THIRTY* E10k's I have sitting around me at the moment and install Linux on it. Or, rather - I'd love to TRY. But I don't have any real notion that any version of Linux, AS IT STANDS RIGHT NOW, will work as well as Solaris on that box.

    Sure, Solaris isn't very user-friendly. GNU/Solaris (Solaris with GNU Tools) is better, but still not anywhere near what most Linux folks are used to when it comes to command-line fun. However, Solaris is *made* to work with Sun hardware. And it does, very well.

    I doubt it highly that someone is going to go buy a US$4M E10k/E15k box and start porting Solaris tools and system utilities *just* so people can run Linux on those systems. Right now, the only reason people have installed Linux at ALL on those systems is for bragging rights.

    If you want to outlay the cash and start-a-porting, I applaud you. I really do. But I won't hold my breath.

    --
    GIR: I'm going to sing the Doom song now. Doom doom doom doom doom doom de-doom doom doom doom doom doom doom...
  47. Sour Grapes by Anonymous Coward · · Score: 0

    I think this is just sour grapes - Sun didn't think of doing this on their Starfires or whatever they call their big(ish) iron, so they're bashing IBM.

    Glenn

  48. They're probably right. by Rebel+Patriot · · Score: 1
    Linux on the mainframe just doesn't compute. Linux isn't designed to run in a virtual machine; implementation decisions that make sense on PC hardware don't fit well in a virtual machine.

    Hate to say it but they're probably right. Linux is 32 bit code made to run on a 32 bit processor. Remember when the 386 came out? DOS ran slower on most 386's, but the new windows ran much faster on a 386 than a 286. While I haven't personally ported linux to a 64-bit processor, it seems likely that you would have more performance issues with the "upgrade." Take for example these benchmarks from tomshardware.

    These dual proccessor motherboards both scored worse in kernel compilations with both processors active! It was faster to run the kernel compilation on a single processor. While two 32-bit CPUs != 1 64-bit CPU, it does illustrate how a major hardware change can make linux (or indeed any OS) flounder around.

    --
    Slackware forever. Honestly, what else would you trust when it absolutely positively has to be stable, secure, and easy
    1. Re:They're probably right. by Phs2501 · · Score: 2, Insightful
      I disagree. It's not as if Linux hasn't been running on 64-bit systems for a long time now - Alpha was the first non-Intel port IIRC. If there's anything that's going to slow down Linux on S/390 it will be how good gcc does on targeting the virtual processor, and how well the IBM devs have tied in device support - not the fact that it's 64-bit.

      With regard to the linked benchmarks, somehow I bet the benchmarkers at Tom's didn't compile parallel. Of course there is not going to be a speed improvement if you don't run a parallel make!

      As someone who builds a small embedded Linux system from scratch (including gcc and glibc), a dual processor system is VERY nice. It cuts down the compile time by at least 30-40%. make -j2 is your friend with two processors.

    2. Re:They're probably right. by bdeclerc · · Score: 1
      Remember when the 386 came out? DOS ran slower on most 386's


      I remember when 386's came out, and DOS did not run slower on them than on 286's.

      You seem to be mixing things up. Windows 95 ran slower on Pentium Pro than on equivalent Pentiums, because 95 still contained a lot of 16-bit stuff, and PPro was heavily optimised to do 32-bit operations.

      Then Intel came out with Pentium-II, which performed as well as PPro at 32-bit, but could reach higher clock speeds, and performed much better at 16-bit code than PPro of equivalent clock speed.

      Linux is now running on both 32-bit and 64-bit CPU's, and there is little reason to assume that it is "performing worse" on 64-bit architectures "because" it was originally designed on a 32-bit basis.

      Kernel compilation is hardly the end-all of tests, and many processes will perform much better on dual-processor systems than on single processor systems, although almost always less than 2x better.
    3. Re:They're probably right. by Anonymous Coward · · Score: 0

      Linux isn't designed to run in a virtual machine; implementation decisions that make sense on PC hardware don't fit well in a virtual machine.

      Maybe not, nut the virtual machine in question is designed to run Linux, so that shouldn't be a problem.

    4. Re:They're probably right. by Wdomburg · · Score: 2

      >These dual proccessor motherboards both scored
      >worse in kernel compilations with both
      >processors active!

      And I'd bet you my next paycheck this is yet another example of the poor benchmarks from Tom's Hardware. They don't mention what options were used, but from the results it's pretty clear they just did "make" instead of "make -j2", so the compile was running a single job at a time.

      >It was faster to run the kernel compilation on a >single processor.

      Let me rephrase that for you - it was faster to run the kernel compilation on a single processor on an single processor kernel, as opposed to running the kernel compilation on a single processor on an SMP kernel.

      >While two 32-bit CPUs != 1 64-bit CPU, it does
      >illustrate how a major hardware change can make >linux (or indeed any OS) flounder around. [

      This is pretty much just speculation though. The proof is in the pudding, as they say.

      And running 64-bit is hardly new to Linux. You have to set the way back machine all the way to late 1994 if you want to glance at the first efforts to make Linux run on a 64-bit platform. So 7 years of the 10 year history of Linux is as a multi-platform 64-bit capable operating system.

      Matt

    5. Re:They're probably right. by crome · · Score: 1

      This is serious crap you are saying. Linux is not 32bit code. As an example take the Alpha, where Linux runs as a full 64bit OS. The Itanium is 64bit, too and various other machines (oh and the S/390 is not 32bit either).

      Moshe Bar

  49. Sun should go to hell by Anonymous Coward · · Score: 0

    Sun is worse than even Microsoft when it comes to this kind of thing. They claim to be embracing Linux but make it very clear that its use is to be limited to end of network solutions and "real systems" such as those based on Solaris should be used for the real work. At least Microsoft's stance on Linux is consistent. Linux on the IBM mainframe is clearly an enormous danger to Sun, whose enterprise servers will be competing in the same space.
    Right now I can see some reasons why people would want Solaris as opposed to Linux for various reasons especially in larger "enterprise" systems. Unfortunately for Sun, those reasons are decreasing on an almost daily basis. At its current rate of progress is would be quite reasonable to assume that the feature set offered by the Linux kernel will surpass that of the Solaris kernel within the next few years. I would love to see Sun either edged out of the market or forced to compete by doing something innovative besides talking trash and suing their competitors.

  50. Legitimate concerns by pamdirac · · Score: 1

    On the one hand, Sun appears to be lashing out at a competitor.

    On the other hand, this article raises some legitimate concerns that IBM will have to address. But they don't have to convince me. They only need to convince their customers.

    As far as I'm concerned, good business or not, running Linux in every possible context makes it stronger.

    --
    John McNair
  51. Nope... by Anonymous Coward · · Score: 0

    Sun sells hardware and service/support contracts. They give the OS away for free. My guess is that they bash linux here because either it actually doesn't run as well on their hardware, or they don't feel like spending time/money training their support people to troubleshoot a foreign OS.

  52. Their strategy by percey · · Score: 2

    There's an article here on cnet news that explains the whole thing. They don't want you to go after the much more powerful and robust linux system that IBM has and go for their version of Linux for smaller systems. The alarming thing is that they are producing their OWN version of Linux, not using Redhat or another company. And also IBM I understand is doing this too. This can't be a good thing for linux at all. Proprietary versions of linux? Or perhaps some people think it'd be okay if its just branded, I just don't think its a good idea at all.
    I think they want to utilize the benefits of linux however they do not want to allow Linux to creep into the larger servers where Sun dominates. And IBM which has AIX 5L (AIX w/linux compatibility) and now a special Linux for the mainframe it directly challenges their most valuable property, solaris which is valuable because all that software is made for it which makes people buy sun systems.
    You find the program you need to run and then look at the systems running it, and unless you're already running AIX or HP-UX your first choice is probably Sun (and sun is usually always a choice). Now Linux comes in, becomes this pervasive server software and Solaris doesn't really look as hot as it did anymore.

    1. Re:Their strategy by Blrfl · · Score: 1
      percey writes:

      The alarming thing is that they are producing their OWN version of Linux, not using Redhat or another company.

      IBM has more in-house OS talent in its pinky than any of the Linux distributors. They know the most about their hardware. So why farm it out to some other vendor who's not likely to have a clue about how to make the OS run on their own platform? (And this is a platform thing. Don't assume the z800 is a big box that's binary-compatible with the x86, because it isn't.)

      It pizzes me off to no end that people have completely forgotten that the whole purpose of an operating system is to provide a consistent API for software to bang against regardless of what hardware it's running on. If IBM's Linux for the 390 provides a properly-functioning interface identical to that provided by every other flavor of Linux, the whole recertification argument becomes a moot point. If something like Apache builds and runs properly on RH for x86, there's no reason it shouldn't on IBM for the 390.

    2. Re:Their strategy by percey · · Score: 1

      I cannot dispute that they have many many talented people working at IBM. I, however, do think that its a negative development that two monoliths are producing their own versions of Linux, if its more than just a branding they run the risk of forking. Besides, there were already working versions of Redhat and I think at least one other distro and now that revenue stream will be gone for those companies. No one will elect to use Redhat's OS/390 version of Linux over IBM's. Its kinda like how ma and pop organizations are effected when a Walmart comes to town. Don't forget IBM also makes many other architectures, and can stake a claim to having unique insight into everything from the AS/400 down to PowerPC systems, I do not see that as a valid excuse for attempting to crush every other vendor in these areas.

      The same goes for Sun, if they make a sparc version of linux, what about the other guys who might be targetting their distros at the sparc architecture? Gone. Its not all bad news, it does mean that linux has come of age in the enterprise server world.

    3. Re:Their strategy by Blrfl · · Score: 1
      percey writes:

      I cannot dispute that they have many many talented people working at IBM. I, however, do think that its a negative development that two monoliths are producing their own versions of Linux, if its more than just a branding they run the risk of forking.

      Possible, but not likely. The only place there's a real difference between platforms is down a few subtrees in the kernel source. With the exception of the C compiler (which has to be tuned per processor) and a few utilities that are specific to one platform (such as hwclock), everything else can be identical. Name it: emacs, grep, Apache, perl, etc. all come from the same set of sources on every flavor of Linux and a number of other OSes, too. If Apache forks, that's a problem for the Apache community, not the Linux community.

      Besides, on the basis of percentage of code developed in-house, RH is just a branding, too.

      Besides, there were already working versions of Redhat and I think at least one other distro and now that revenue stream will be gone for those companies. No one will elect to use Redhat's OS/390 version of Linux over IBM's.

      I'll give you three guesses who did most of the work getting the kernel to run on the 390 (three letters, very close to H-A-L). If the distributors have their acts together, then their build process is fully automated and retargeting on a platform shouldn't be much more than telling the C compiler what CPU it'll be cross-compiling to.

      Its kinda like how ma and pop organizations are effected when a Walmart comes to town.

      That's a whole different ball of wax. Wal Mart and mom-and-pop shops sell the exact same roll of toilet paper sold to them by a third party. Wal Mart's advantage is that it buys them in such huge quantities that it gets a deep discount that the M-and-P shops can't match. IBM actually engineers and manufactures products, and for someone to buy any flavor of 390 Linux, IBM must get another of its machines out into the world.

      That's not the case for RH, who manufactures a product that runs on commodity Intel boxes which are made by thousands of different manufacturers. RH is actually has a competitive advantage here because they can adapt their product to any platform they see as being ubiquitous enough to be a commodity and therefore a good revenue stream with comparatively minimal effort compared to what a hardware vendor must do before a new platform sees the light of day.

      Don't forget IBM also makes many other architectures, and can stake a claim to having unique insight into everything from the AS/400 down to PowerPC systems, I do not see that as a valid excuse for attempting to crush every other vendor in these areas.

      The Intel PC is in IBM's product line, too, but you'll notice that they don't sell an IBM-branded Linux for that, either. If IBM's distributions of Linux for their own products superior to anything else, why shouldn't they win? Nobody screamed bloody murder when HP jumped on the bandwagon.

      RH is arguably the most successful shrink-wrapped Linux for Intel on the planet, but I hear very little hew and cry about the small distributors who've gone tits up because of that success.

      Lots of times the big guys win, and every once in awhile one of the little guys does something great that propels them into being big. Sun started its life as one of the little guys, too.

      Its not all bad news, it does mean that linux has come of age in the enterprise server world.

      Yes, it does. And that coming of age means big businesses like IBM, Sun, HP and others will get involved. But it's a free market, and that's how the free market works.

    4. Re:Their strategy by Wdomburg · · Score: 2

      >The alarming thing is that they are producing
      >their OWN version of Linux, not using Redhat or
      >another company. And also IBM I understand is
      >doing this too.

      IBM has actually done a commendable job of keeping their support for Linux as vendor agnostic as possible, while still working with Red Hat, SuSE and TurboLinux for more comprehensive services.

      Though, not being an IBM executive, I cannot rule out the possibility of a "proprietary" distribution, I don't see that it would be in IBM's interest. There is a signifigant cost investment to be made in supporting an operating system, and recouping that investment when other distributions are available would be nigh impossible.

      >This can't be a good thing for linux at all.
      >Proprietary versions of linux?

      The first question here is how exactly you'd make it propritary. The kernel, core libraries, utilities and daemons are all under the GPL.

      You can certainly write your own versions of libraries and utilities. And it's nigh trivial to make a compatibility layer in a proprietary Unix kernel. But isn't that what AIX 5L is?

      And then you have to wonder how you're going to get any return on investment. Competitive pricing in the Linux world is free. Sure you can make money off support, but you don't need to create a distribution for that.

      Even if the pricing isn't a consideration, what reason would there be from a customer standpoint to use a proprietary version? Half of the draw of open source is in its openness - no vendor lock in, no abandonware, the ability to do in house modifications and support, no licence management overhead.

      I'm not sure why Sun would be pursuing a custom version of Linux either, and their press releases never say that they are. The only thing that could possibly be interpreted as this is "Sun will ship for the first time a full mplementation of Linux on a new line of general-purpose servers."

      Howver, that makes no mention of it being their own version of Linux; just that it will be a full implementation. This is likely to differentiate it from their line of Linux powered server appliances (aquired from Cobalt).

      Until they release further details next quarter, its pretty much impossible to do anything but speculate.

      Matt

  53. screw Them by Anonymous Coward · · Score: 0

    Is that how it goes? The one minute they support linux and the other they bash it?
    I think it is time to start learning C#. Java is slower than a mule anyways.

  54. Bashing is pointless. by wo1verin3 · · Score: 1

    Like all the other alternatives. Linux is a solution. It's not the solution for everything, it's not the answer to lifes problems, but it's a solution in some cases.

    I think Sun's is WAY out of line making such a broad and unqualified statement.

  55. Ok... by Warthog9 · · Score: 1

    So if sun doesn't think that the VM in the Linux kernel isn't for Mainframes, I don't see them making a good suggestion for how to improve it, but just bashing the current philosophy. If you are going to crticize, at least make it constructive Sun.

    As for my thoughts on the issue directly and against Sun, if linux runs on a mainframe people will find the problems that exsist for linux on the platform and begin striding towards correcting those, the same way with any problem (I'll take the StrongARM processor that my linux iPAQ runs) it's not perfect but it's being worked on and thats why linux keeps going because people go "hey lets try and make this work on here because it would make job X easier" and they do it..... now if only I could re-program my toaster.

  56. Hey - me too! by SubtleNuance · · Score: 5, Funny

    And it was a seriously easy. There were never a problem with the virtual machine suddenly giving up the ghost from underneath us, and we'd never loose a Samba process for any reason whatsoever. We had load averages that were very normal, and it was like we were always scrambling to find new reasons to use it, as opposed to setting it up and just having forgetting about it. We used to tell the students that the machine was seriously hot, and it would never fall-over. We were even thinking of doing up artwork -- AND STARTING A RELIGION!


    All those impressive demos where they have 32 hojillion instances of linux running on a mainframe are so meaningfull. Sure, you can do without it, but it will do everything. If you try actually working with the setup, you never have to reboot your machine -- instead of doing it 10 times a day, and those other take forever to freakin' reboot.

    But who are you going to believe, ummm, me or ummmmm, that other guy.

    Sheesh.

    1. Re:Hey - me too! by Anonymous Coward · · Score: 0

      > and we'd never loose a Samba process for any reason

      I'm all for freedom of Samba processes too, but did you every _lose_ one, by chance?

    2. Re:Hey - me too! by Anonymous Coward · · Score: 0

      > but did you every _lose_ one, by chance

      Hmm.. did you _ever_ lose one yourself?

    3. Re:Hey - me too! by Anonymous Coward · · Score: 0

      The other guy. As we've learned in the past two weeks, Canadians tend to whine until they get their way, and when things go really badly make delusional accusations of a vast US media conspiracy.

    4. Re:Hey - me too! by westyx · · Score: 1

      The other guy - lower slashdot id#. ... what - you think we're gonna think about this???

  57. Slashdotters read this by Anonymous Coward · · Score: 0

    Repeat 10 times: "I DONT CARE. I DONT OWN ONE OF THOSE. NEIGTHER DO YOU."

    Good. Now the mainframe propblem, if there is a problem, is gone.

  58. Sun is just mad.. by gh0ul · · Score: 1

    Sun Microsystems is just mad because IBM made it there before they did?

  59. Not really bashing linux by lkaos · · Score: 1, Redundant
    Perhaps I do not have as finely tuned of an anti-linux ear as the /. admins but it doesn't really seem to me that Sun is bashing Linux.

    Sun brings up a few very valid points:
    1. Linux on mainframes is still running on top of a properitery VM. This kills the whole Free Software argument for using Linux.
    2. Linux is not designed to run on top of z/VM. It isn't :) That's not a bad thing, it's the truth.
    3. Using Linux on top of z/VM limits the amount of fault recovery that can occur compared to IBM's old z/OS. This seems logically to me.

    I don't think Sun is trashing Linux. I just think they are saying that Linux does not scale to Mainframes that great right now. Perhaps the only thing they aren't doing is saying that while there are issues with Linux now, the fact that Linux is FS allows those issues to be resolved in the future. Of course, why would a company endorse another company's product?
    --
    int func(int a);
    func((b += 3, b));
    1. Re:Not really bashing linux by crush · · Score: 1

      Linux on mainframes is still running on top of a properitery VM. This kills the whole Free Software argument for using Linux.

      I appreciate most of what you're saying, but it's just not true that GNU/Linux is running on top of a VM on mainframes. Yes, that is the case that Khan is criticising here, but there is the option of running GNU/Linux natively without the VM. This had already been addressed by Caiman, Think Blue, SuSE, LinuxKorea and others and the installation on big iron is explicitly discussed in IBM's RedBook on the subject.

      It seems to me that all that Khan is done has to throw into question the idea that "high availability" is not a problem with a "server farm in a box". But this was not explicitly argued as being the major advantage for the multiple image VM-hosting by IBM. They argued that it allowed ease of administration.

    2. Re:Not really bashing linux by lkaos · · Score: 2

      I took that from what the article had stated.

      He obviously doesn't address what you point out, but I think the argument against the VM-ized version is valid. The benefits of Free Software obviously do not apply when the underlying system is closed.

      I wonder if Linux still suffers the horrible performance issues when running natively verses when running on top of a VM (or atleast, the issues pointed out in the article).

      --
      int func(int a);
      func((b += 3, b));
    3. Re:Not really bashing linux by crush · · Score: 2

      He obviously doesn't address what you point out, but I think the argument against the VM-ized version is valid. The benefits of Free Software obviously do not apply when the underlying system is closed.

      On the other hand some people will tell you that a small in-house team of hire'n'fire developers can make better progress on some things (like Intel's C compiler) than "Open Source" projects. Contrary to the mantra that "Open" development is always better for performance issues the advantage of Free Software have to do with Freedom: improved performance is a frequent desirable epiphenomenon.

      I wonder if Linux still suffers the horrible performance issues when running natively verses when running on top of a VM (or atleast, the issues pointed out in the article).

      The same general problem would still occur. However LPARed S/390 definitely doesn't suffer the same performance hits as you would expect.

      Ps: Before anyone flames me I consider Free Software to be wonderful and desirable, but I don't believe that it is always more efficient or accurate or perfect as a result of that development process: it /can/ be, but it is not an immutable outcome

    4. Re:Not really bashing linux by lkaos · · Score: 1

      Contrary to the mantra that "Open" development is always better for performance issues the advantage of Free Software have to do with Freedom: improved performance is a frequent desirable epiphenomenon.

      I didn't really mean to state whether or not the benefits of FS are real, just that they cannot exist when a majority of the system is closed.

      As far as I know, not many advocates of FS claim FS is better because it is faster. FS on the average is not faster because either certain optimizations are copyrighted (Intel C++ for example).

      --
      int func(int a);
      func((b += 3, b));
  60. uh by poemofatic · · Score: 1

    by VM they mean the virtual machine in which linux has to run. Not linux's virtual memory system.

    And they seem to make some good points about this setup not being optimal -- i.e. you don't get IBM's legendary stability, etc. by running linux in a VM. The only obvious FUD I picked up on was the business of applications needing to be certified before they would run. Since I'm assuming that the only *technical* reason people would run linux on an ibm mainframe is to use the apps.

    Any IBM hackers care to comment? Anyone run linux on a mainframe?

    --

    When in doubt, have a man come through a door with a gun in his hand.

  61. Guy's got his facts wrong. by Blrfl · · Score: 2, Informative
    NitsujTPU writes:

    If the article is correct, then the z800's running zVM emulate Intel x86 architecture in order to run Linux. Heck, even poorly written native compiled code generally has advantages over such a set up.

    That's not what IBM is doing at all. The version of Linux that runs on the 390 platform is natively compiled for that system and has been tuned to work with it. This is no different than Linux on any of the dozen or so other platforms it's been designed to run on.

    Sun's FUD-slinger got his facts wrong more than half the time in that article, which in most places would get him grade of "F". I've been using and buying Suns for 16 years, and while I really like their products, I hope IBM takes every opportunity to point out what a nitwit he is.

    The moral: Just 'cause it says "Linux" on the label doesn't mean it's running on an x86.

  62. FUD. by Jay · · Score: 2, Insightful

    He basically has 1 valid point. Linux under VM is not as well behaved as it should be, for exactly the reason mentioned. Coming from the PC world where all hardware is real, it treats all memory as if there isn't anything else better to do that use it all for file buffers. Under VM, a better plan is to check to see if there is any memory pressure being exerted on the machine before using your entire memory allocation for buffering.

    They're working on that.

    As for the rest, it's mostly FUD. The endian-ness is not an issue for 95% (wild ass guess) of apps that I have seen. Maybe except for DB2. You have to plan your maximum capacity in a discrete server farm just like you do in a virtual one. You also get capacity upgrade on demand with a phone call with the IBM hardware. They dont even have to send out a CE to do anything. Let's see SUN do that.

    You wouldn't want to use it as a compute farm, but as a database server or news server or something which is usually I/O bound. They ain't exactly ferraris, more like 18 wheeler big rigs.

    --
    You think emacs is evil?! You've never used VM's XEDIT have you?!! That's evil, baby!
    1. Re:FUD. by Improv · · Score: 2

      Well, actually, I think they had one good point
      and one bad/silly point.. the good point being
      the relative cost-effectiveness being lower on
      the mainframe (although I wonder if they assume
      I/O or CPU is more important a measure of effectiveness)
      The bad/silly point was endianness -- apparently
      the mainframe is not little endian. It's amusing
      they didn't mention the name of the other kind,
      big-endian. If I recall correctly, SUN systems are
      also big-endian. And of course, I agree with you
      that endianness really is a non-issue.

      --
      For every problem, there is at least one solution that is simple, neat, and wrong.
  63. Design by Trillian_Angel · · Score: 3, Insightful

    As far as I know, Linux wasn't really designed to be a mainframe anyway... so whats the big deal really? Linux is a web server/ server system, great for apache and the likes, stable and a life saver for small companies. So Sun has the better equipment and software for mainframe. IF linux had been designed to do that then there would be a real contraversy in the situation anyway. Its an interesting article with lots of good points, but its like comparing tomatoes and oranges.

    --
    -- RJ
    1. Re:Design by sydneyfong · · Score: 1

      As far as I know, Linux wasn't really designed to be a mainframe anyway...

      No, Linux was not for anything specific. Sure it wasn't really meant to be used on mainframes, but it can. Similarly, it wasn't designed to be used on embed devices, but it can. Heck, it wasn't designed to run on desktops at first, but now everybody is trying to push it in that direction.

      Linux is a web server/ server system, great for apache and the likes, stable and a life saver for small companies.

      Oh yeah? For me it's also a great desktop and a great programming environment. Everybody may be using Linux in his/her way. Afterall, that's the whole point of linux, isn't it?

      --
      Don't quote me on this.
  64. Goes to show... by clump · · Score: 3, Informative
    Linux isn't designed to run in a virtual machine;

    Linux isn't designed at all, which is good. Thats why its so flexible. There was already that debate a while back.
    Applications that run on Linux for Intel need to be recompiled and recertified for each new platform; thus the application portfolio to run Linux on a mainframe is small(9).

    Oh please. Like your going to have an easier time compiling non-Linux software? Still think so given how open and portable most Linux software is? Is mainframe software as portable? Is there lots of free mainframe software to port? Thats almost as irrational as Microsoft's "Linux isn't free" TCO argument. Per that can-of-worms, because both systems have TCOs means NT itself *is* free?

    Articles like this are interesting because Sun definitly has a conflict of interest with Linux. They need to appear as if they support it so new blood will buy SPARC hardware with Solaris, but they also don't want people 'liking' Linux over Solaris/SPARC.

    Personally, I love Linux on SPARC. I would prefer Sun making Linux more 'Enterprise'-like instead of hawking Solaris as a big-brother. However, I understand that Solaris is a huge investment and one they probably will think is superior for years to come.

    For their sake, I hope the Penguins don't squish them. But if they don't look both ways before crossing the street...

    1. Re:Goes to show... by devinoni · · Score: 1
      My favorite quote from the article is:
      Often the difference in Intel versus mainframe applications makes porting difficult(10).
      (10) Intel uses something known as little endian; a mainframe uses something different. This is significant for certain applications and makes the port difficult.
      Endianness hasn't stopped people from porting application to SUN's big-endian sparcs.
    2. Re:Goes to show... by Octorian · · Score: 1

      I think the intent isn't big-endian vs little-endian. The issue is that the way mainframes store numbers is "different" than normal machines. Kinda in a weird way, though I'm not up to speed on the details.

    3. Re:Goes to show... by Octorian · · Score: 1

      Why does everyone think Linux on Sparc would be better? What you like about Linux isn't the kernel, but rather the userspace apps and utilities. Well, run them on Solaris! Actually, Sun does seem to be progressively improving Solaris with regard to userspace stuff. Things just move slower in their world in that area, since the core of the OS and the hardware are a bit more important.

      (posted from a Sun Ultra 30 running Solaris 8, KDE 2.1, and browsing with Opera 5)

    4. Re:Goes to show... by devinoni · · Score: 1
      To quote this http://www-1.ibm.com/servers/esdd/articles/linux_s 390/:
      S/390 is a big endian system. Any code that processes byte-oriented data that originated on a little endian system may need some byte-swapping. The data may have to be regenerated or, if that isn't possible (for example, shared files), the application may have to be reworked to adjust for processing little endian data.
    5. Re:Goes to show... by Tony-A · · Score: 2

      Say you have a value 12345678 in hex
      Little Endian : addressed by least-significant byte
      78 56 34 12
      Big Endian : addressed by most-significant byte
      12 34 56 78

      You have a value 00000001 in hex stored at location 1000
      32-bit value at location 1000 both endians.
      16-bit value at location 1000 little endian.
      8-bit value at location 1000 little endian.
      16-bit value at location 1002 big endian.
      8-bit value at location 1003 big endian.

      Big Endian:
      Numeric order corresponds to character order.

      Little Endian:
      Numeric order unrelated to character order.

    6. Re:Goes to show... by clump · · Score: 2
      What you like about Linux isn't the kernel, but rather the userspace apps and utilities.
      Well you see, not really. Though the userspace apps are nice, so is the kernel. The Linux kernel runs on a large amount of architectures, including the SPARCs. Also, Linux is Free in all senses of the word.

      Not that Solaris isn't nice, but why run a single OS for a single arch? If Solaris can do almost everything Linux can do, why not just run Linux?
    7. Re:Goes to show... by shani · · Score: 1

      Why does everyone think Linux on Sparc would be better?

      As a software developer, the reason I would prefer Linux on Sparc is the same reason I prefer open source in general: I can see what's really going on and fix it if necessary.

      We added Sun's Recommended Patches to our servers in December 2001, and one of our applications started leaking memory. I spent a few days and finally tracked it down to the gethostbyname_r() call in POSIX threaded applications in the patched Solaris. We sent Sun support a 20-line program to demonstrate the problem.

      Well, it's now almost March 2002, and Sun hasn't fixed it. If it was a Linux box, I probably would have just fixed it myself.

      If you're forced to use expensive, slow Sun hardware and deal with their incompetent support staff, at least you can fix the problems and serve your customers with Linux!

  65. Sun has lost... by HogGeek · · Score: 1, Flamebait

    ... Their competative edge. With the hardware issues on the Sparc II modules that affected companies such as e-bay (and the company that I work for), they have given alot of their customers black eyes. We are converting from a SUN E10K to IBM RS/6000's running AIX and Linux, because our clients can no longer tolerate the Sun crashing...

    Sun is just trying to cover the ass, as they are losing a lot of business to IBM.

    1. Re:Sun has lost... by Anonymous Coward · · Score: 0

      There seems to be some misunderstanding here as to what exactly an "IBM mainframe" is. An IBM mainframe is an archaic, clunky, proprietary mess weighted down by almost 40 years of useless backward-compatibility, which happens to cost in the area of millions of dollars per year to lease and operate. An RS/6000 running Linux or an AS/400 for that matter is not a "mainframe" in the context of the Sun article. The economics of the Sun article are on target, IMHO.

  66. Why with VM by kichiguy · · Score: 1

    Pardon my ignornace, but can't you run Linux directly on IBM mainframe iron without VM between you and the hardware? How is the Linux solution in that case?

    1. Re:Why with VM by PoiBoy · · Score: 1

      You can -- but then you've got one box running one linux kernel. The point of using a VM layer is that you can then have a bunch of smaller servers all in one nice box, and IBM's selling this idea mainly to emphasize the point that it's easier to take care of one big box than having to chase 100's of smaller boxes scattered all over BFE.

      --
      Sig (appended to the end of comments you post, 120 chars)
    2. Re:Why with VM by Anonymous Coward · · Score: 0

      The virtual architecture is not identical to real architecture. There are some differences. Also you have those hypervisor calls (sort of like system calls) that you can invoke if you don't want to deal with all of those messy hardware details. If you use those then you can't run native. I suspect you will find those used in some of the drivers written to port Linux to the mainframe, especially if the function is already available in the hypervisor.

  67. Good marketing exec by JaBean · · Score: 1

    I don't know about ya'll, but he's got me convinced. Bow to the powers of coercive reviews ...and give me Solaris!

    1. Re:Good marketing exec by Anonymous Coward · · Score: 0

      Why don't you just continue wanking? That's what you do for the most of your time, isn't that right?

  68. Of Course not! by Zelet · · Score: 0

    That is what I said and I am still at a 0. :( It is okay I suppose, I was a little vauge since it was a while ago that I heard about it.

    --
    ...And when they came for me, there was no one left to speak out for me." - Martin Niemoeller (1892-1984)
  69. Stop the presses: Sun dislikes IBM solution! by Anonymous Coward · · Score: 0

    First you ask Bill Joy what he thinks of C#, and now this. A new low for Slashdot.

  70. What do we think?!! by garoush · · Score: 2

    "What do the folks on Slashdot think?"

    This is easy: anything posted on /. that has to do with Linux or Windowz will give you tones and tones of posting.

    --

    Karma stuck at 50? Add 2-5 inches.. err.. 2-5x Karmas Count to your pen1es.. err.. Karma all naturally and private
  71. Wrong question. by Anonymous Coward · · Score: 0

    Why ask the average slashdot reader what he/she thinks?
    The average slashdot reader is not in the market for a 400k$ server.

  72. Re:Sun is not Linux's friend -Irrelevant! by Anonymous Coward · · Score: 1, Insightful

    Friend or not, in the long run Sun will need Linux for Sun just to survive. So the question isn't is Sun sincere in their Linux efforts nor does this attack on IBM signal new antagonism between Sun's interests and Linux, but instead do we even need to notice Sun when it says things like this?
    The answer is no, of course not.
    We neither plead with Sun to get onboard with linux nor get angry and bothered by this kind of attack. It's 100% irrelevant.
    It's up to Sun Microsystems to come to Linux and save itself; nothing we do or say can positively influence their intransigence/willingness, so we're best off ignoring them completely no matter what they say.
    Think how absurd it is for Linux people to get upset or harbor paranoia about Sun!
    Without doing anything hostile to target or hinder Sun, or Sparc or Solaris, Linux has compelled Sun to include linux compatibility runtime, to offer an official JDK for Linux, and to acquire a Linux on X86 hardware business and now, to expand it into the midrange of server offerings.
    Linux is an irresistible force and, no matter what some dinosaurs over there may think, SUN is not an immovable object, it will orbit Linux or it will fall into Linux and burn up. Whether they survive the journey or not they will come. They are coming over even now.

  73. How do you tell when a marketeer is lying? by Dammital · · Score: 3, Insightful
    Yeah, you know the answer to that one, don't you?

    This piece was so full of FUD that I could scarcely believe it.

    z/VM is a niche operating system with virtual machine (VM) support for new hardware features added late or often not at all(3)

    Then why did IBM have to port the TCPIP stack from the VM world into MVS, if VM is so far behind?
    This is Linux. It's designed for Intel. It's not tuned for [S/390].

    Linux is "designed for Intel"??? What about M68K, PPC, ARM, Alpha, SPARC and others? (See the Debian ports page for a more complete list.)
    The "legendary" IBM S/390 [reliability] IBM references are the result of decades of development work on IBM's flagship mainframe operating system, known today as z/OS.

    Yes, MVS (z/OS) is rock solid reliable. But the machines don't bust either. CPU recovery has been an integral part of the architecture for almost 30 years. If a processor breaks, another takes over with no application effect, or a spare is assigned. Someone on the ibm-main list today mentioned that the processors are themselves duplicated on chip, with comparison logic to ensure that both sides are computing the same thing. Does Intel even parity-check their processors?
    thus the application portfolio to run Linux on a mainframe is small

    Small? Install a copy of SuSE SLES in a S/390 LPAR (logical partition, a hardware implementation of VM that is delivered on EVERY S/390... no z/VM necessary) and see how much software was delivered with it. You wanted OpenSSH and OpenSSL, though SuSE didn't deliver it? Go to the web, download it, and do configure, make, make install. The big problem with application portability is the proprietary vendors that ship binaries only.
    the difference in Intel versus mainframe applications [WRT endian-ness] makes porting difficult

    What an amazing assertion. Wish Khan had provided a reference.
    Why put an open operating system such as Linux on a closed proprietary mainframe?

    Merde. Why run it on a closed proprietary SPARCstation? Or a closed proprietary Mac?

    Khan makes a couple of decent points, particularly regarding z/VM skills. But the hyperbole is way out there, and it's hard to take him seriously.

    1. Re:How do you tell when a marketeer is lying? by shani · · Score: 1

      Someone on the ibm-main list today mentioned that the processors are themselves duplicated on chip, with comparison logic to ensure that both sides are computing the same thing. Does Intel even parity-check their processors?

      I'm not sure how one would parity-check a processor per se, but the Itaniums do indeed parity check their on-chip cache memory.

      And remember, we're bashing Sun today, not Intel. Sun makes their own CPU's. Yes, they might be expensive, but they sure are slow!

  74. The struggle for hearts and minds by ynotds · · Score: 2, Interesting

    I fondly remember VM as the first operating system I ran into that embodied a really good idea.

    There was a stage during the '80s when I was working more as an industry analyst than as a developer when Sun and IBM between them had become two of the then only four serious pillars on which the future of computing rested at a conceptual level.

    At that level, Sun was the embodiment of Unix, taking evnagelical responsibility for the cause, and it is reasonable to assume that within their own envorins they genuinely see themselves in that position still.

    From my own biased perspective, I felt they abdicated that authority when they allowed their elegant Network-extensible Window System (NeWS) to be rolled by a tide of industry resistance that mobilised against the upstart Sun and behind the then clearly inferior X.

    But I'm sure in Sun's hearts they still believe they are the ultimate repository of deep understanding on all things Unix and are being genuine and honest in the technical basis for this critique.

    The real problem is thay they can't see beyond their own world view. They do not have places in their heart for deep understanding of either the VM nor the Linux view or the world, let alone the two in combination.

    Still Sun struggles to find its own identity and focus, to say nothing of a sustainable business model for the future.

    From NFS to RISC, to industrial strength Web servers and on to Java, Sun has been a major contributor to the direction of mainstream computing, but now seems to be edging closer to following the fall to oblivion of that other former pillar of hearts and minds, Digital.

    It will be a worse than sad day when we finally have to convey Sun to history, especially if that comes before Java gets to really stand on its own feet.

    --
    -- Our systemic servants do not good masters make.
  75. Please ..... by Anonymous Coward · · Score: 0

    why would you be a linux advocate?

    Wouldn't it make sense to be an advocate of quality computing ? Or better yet and advocate of ROI, or better yet an advocate of quality research.

    You go apply for a job and put "linux advocate" on your resume and I will put advocate of "quality computing" on mine. Guess who will look like the person who has nothing better to do that bash anything that isn't linux because they think MS has actually harmed them in some way ?

    Come on folks. Do you think Sun, IBM or even MS is an enemy of the state or is "out to get you".

    You linux advocate freaked because Media players caches it's requests and CD info. You ever look at how much info linux keeps, try everything and I don't think it informs you of all the info it keeps on the outside of a "insert any linux company name here" box.

    Please get a life and move on to something that matters.

    btw I have run linux on a ibm vm. Question is there is no real advantage to. I can do everything linux can on Solaris only in a bigger fashion and I get better more scalable system with a better support organization.

  76. z/VM is *not* x86 emulation... by Noel · · Score: 4, Informative
    People these days seem to forget about the overhead of interpretters and virtual machines. If the article is correct, then the z800's running zVM emulate Intel x86 architecture in order to run Linux.

    Sorry, but z/VM has nothing to do with emulation. z/VM is a low-level system that simply (or not so simply ;-) virtualizes the hardware by providing one or more virtual machines, each of which can run any native OS. As far as the client OS knows, it's running on the bare hardware. The z/VM layer provides the ability to flexibly divide the hardware resources between the VMs, and guarantees that each VM is completely isolated from all other VMs. In the case of Linux/390, the Linux kernel and applications have been compiled to run natively on the S/390 architecture. Check out this Linux for S/390 FAQ for more info.

    1. Re:z/VM is *not* x86 emulation... by Chemical · · Score: 1

      This is same thing as Logical Partitoning (LPAR), right? I've never worked with any of IBM's zSeries (S/390) servers, or Linux on an IBM mainframe, but I do know a little bit about iSeries (AS/400). My company has one 400 LPARed into two systems each running a seperate installation of OS/400, running different applications, and each having their own DASD. One thing I always wondered about this is when you LPAR a system, when one partition is running a highly CPU intensive job, does it eat into the other system's CPU, or does each system have a set ammount of CPU time it can use? If the prior, it seems like kind of a crappy idea, as if both systems were trying to run CPU heavy jobs at the same time it would slow them both to a halt. If the latter, it still seems like a crappy idea as neither system would be able to use the full potental of the machine. How exactally does logical partitioning deal with this?

  77. Marketing as Op/Ed by Alethes · · Score: 1

    The latest marketing gimmick that seems to be catching on is selling products by providing you with a trusted, informed, and seemingly unbiased opinion. Can you really trust the biggest competitors' opinions of anything? Sure, they may have valid points, but what are they *not* pointing out? IBM probably has just as many reasons as to why Sun's implementation is clunky, illogical, etc., but it really doesn't matter. What matters is if the tools meet the specifications that you, the consumer need. Maybe we can hurry and give this type of marketing a bad label so the suits can move on with their next gimmick.

  78. Mainframe Staffing..... by jsimon12 · · Score: 1

    The article states that there are problems finding mainframe staffing? I think this is totally unfounded. Many Many orginizations already have large mainframe implementations and want to roll the applications to midrange size systems. Converting to Linux would allow them to port the application but still use legacy hardware and legacy hardware resources (take a company like EDS for example, they have tons of mainframe operators already). So there is a vertain leveraging of resources (did I say leveraging?). But then again is Linux really a good mature midrange platform? And do you still really want to be tied to IBM, ie Big Brother, ie they may be cheap up front but they ALWAYS bill ya in the end.

  79. Reading between the lines... by Pig+Hogger · · Score: 3, Funny

    There is a lot of activity around Linux in the marketplace today. Strategies differ from vendor to vendor, and there are a variety of ways companies are experimenting with, evaluating, and deploying Linux in their IT infrastructures. Just like any technology decision, if and where to deploy Linux is a substantial decision which involves matching the right technology to the right business problem.

    READ: Of course, with an infinite amount of money, every business problem is solvable at the highest profit for the problem-solver...

    Just as it is important to understand when technology can provide you with an advantage, it is equally valuable to know when a technology is not suited to a particular task regardless of how 'hot' that technology might be. Sun does see a place for Linux in the IT infrastructure, as evidenced by our recent announcement to ship Linux-based servers. Sun introduced low-priced, horizontal Linux servers as an alternative to proprietary systems. Suited for Web delivery at the edge of the network, Linux servers offer an alternative to proprietary and closed environments such as Microsoft Windows.

    READ: But Sun is of the opinion that it's proprietary servers, running with proprietary hardware (such as the special disk drives with the "magic" secret partition table, and the memory chips with the notch moved 1/10 inch) and, of course, running proprietary Solaris Operating System is the best solution overall, especially if you have an infinite amount of money to throw at YOUR problem...

    Recently, IBM announced a new 'Linux-only' mainframe, the z800, which IBM is promoting as a way to consolidate multiple Linux and Unix[r] servers(1). Running Linux on a mainframe doesn't change the fact that you must still maintain an expensive, proprietary system, defeating the whole purpose of introducing open standards like Linux. Although it's technically possible to configure such a system, the question remains, "How well-suited is the system to the task?"

    READ: Definitely, here, we have a blatant attempt at fitting an hexagonal peg into a pentagonal hole. Linux was conceived to run on discarded low-end hardware in order to satisfy teenage-geek impulses, which is quite a different thing than to run on the Big Iron dinosaurs IBM is well-known for.

    Linux on the mainframe just doesn't compute. Here's why:

    READ: You just can't run JCL and CICS and MVS and CMS and Assembler on Linux. Cobol (even GNU-Cobol) will make the kernel break into hysterics.

    Linux on the mainframe is actually hosted by another proprietary operating system, z/VM. The optimized operating system for IBM mainframes is z/OS, formerly called MVS(2). Compared to z/OS, z/VM is a niche operating system with virtual machine (VM) support for new hardware features added late or often not at all(3). And Linux isn't designed to run in a virtual machine; implementation decisions that make sense on PC hardware don't fit well in a virtual machine(4). This is Linux. It's designed for Intel. It's not tuned for the mainframe hardware in which it's running.

    READ: On the other hand, Solaris is designed to run (well) solely on Sparc architecture that made Sun famous.

    Linux on the mainframe is complicated; this isn't Linux running on a two-way Intel server. Despite IBM's claims of easy management(5), customers still need a special machine room and specially trained staff for both z/VM and Linux. Finding mainframe staffing is an obstacle in many organizations(6); combining mainframe and Linux staffing further complicates the matter. Running multiple Linux images still requires administration that needs to grow with the number of images being run.

    READ: Of course, you don't have any of those virtual machine nonsense with Sun products: you simply plug in as many boxes as you need of machines. No more software headaches for your operators!!!

    Linux on the mainframe can't respond to the workload demands of Web serving with high utilization--something IBM touted at the time of its z800 announcement. Horizontally scaled Linux farms are designed to handle unpredictable demand with above average peak loads. As demand rises, a load balancer distributes the traffic evenly across servers, which increases utilization. Because design capacity needs to handle peak demand, server farms often have a low utilization.

    READ: Why do in software what can be done far more profitably (and piracy-immune!) in juicy, expensi^h^h^h^h^h^h^h profitable hardware?

    Given the relatively low cost of hardware, some organizations find this trade-off acceptable to ensure appropriate service levels. Contrary to what many believe, consolidating a Linux farm into multiple images on a mainframe would not change the demand pattern. Although z/VM can start and stop Linux images, it cannot dynamically add resources to match demand. As a result, a mainframe would need to size for peak demand just as the Linux farm would; high utilization is a myth.

    READ: Of course, you can dynamically add and remove ressources from a Sun cluster with the use of specially trained monkeys and servoids, which is a better proposition than software hocus-pocusery within the deep, dark bowels of an IBM mainframe.

    It's neither fish nor fowl. Linux on the "mainframe" is not an open system, and there is little incremental RAS benefit. Although IBM claims "zSeries servers inherit the legendary IBM S/390 strengths in the areas of fault avoidance and tolerance, recovery from failures, and concurrent maintenance and repair for "always-on" availability"(7). We don't believe this to be true for zSeries servers running Linux. The "legendary" IBM S/390 strengths IBM references are the result of decades of development work on IBM's flagship mainframe operating system, known today as z/OS. The fault recovery features of z/OS are not found in Linux. z/VM does have some fault recovery features, but it is not nearly as resilient as z/OS. For example, z/VM cannot take advantage of Parallel Sysplex clustering, and VM hypervisor is an added single point of failure(8).

    READ: Of course, Sun cannot take advantage of Parallel Sysplex clustering either, but that's more alphabet soup to muddy the waters and instill doubts into potential buyers.

    Applications that run on Linux for Intel need to be recompiled and recertified for each new platform; thus the application portfolio to run Linux on a mainframe is small(9). Consolidation without application availability just can't happen--and if the applications don't run on your platform, or if there are costly ports and changes to be made, cost savings can't be realized. Often the difference in Intel versus mainframe applications makes porting difficult(10). Additionally, different applications are ported to different distributions of Linux (for example, Red Hat, SuSE, and Turbolinux). Getting applications to run on the mainframe might require supporting multiple distributions of the 'same' Linux operating system.

    READ: Again, Linux fragmentation is a terrible tragedy that will never happen to Solaris.

    The economics just don't work. IBM claims it is financially justifiable to consolidate as few as 20 Linux servers on a z800(11). With an estimated starting price of $400,000 for a z800(12) with a single CPU engine enabled, that claim seems exaggerated compared to Linux servers that hover in the $1,000 to $2,000 range. Sun's rack-optimized 1U form factor servers start at list prices under $1,000. When customers realize Linux on mainframe utilization will be low, and administration costs have still not been factored in, you can begin to see how the costs will add up. And let's not forget the support costs that will need to be purchased, either from the distributor or IBM Global Services. One example of a distributor's cost on an IBM Multiprise ran in the tens of thousands for the initial services and thousands a month for ongoing service(13).

    READ: Even if " nobody ever got fired for buying IBM ", we whish the same could be said about Sun.

    Thus, when considering consolidation projects, why consider putting workloads onto a completely different and more expensive architecture? Why put an open operating system such as Linux on a closed proprietary mainframe? Why consolidate on a system with limited application support and one that demands a rare combination of skills?

    READ: IBM is double-plus uncool amongst geeks. SUN is the hip thing to use in IT!!!

  80. You can have access to a mainframe running Linux.. by (H)elix1 · · Score: 2

    As you wish! Check out http://www-1.ibm.com/servers/eserver/zseries/os/li nux/lcds/index.html and you can sign up for a chuck of a linux powered big iron... They even toss in some of their software. Very cool.

  81. Ahh by NitsujTPU · · Score: 1

    Ahh, that makes sense. Sorry, I am now familiar with what is going on. The Z800 is being multiplexed in the VM. DUH! Yeah, that does actually make some sense! Now I am feeling quite foolish. I remember reading of this system before. My apologies. Please mod the above comment down. This technology DOES make sense, because it would scale better than merely purchasing multiple rack units.

    1. Re:Ahh by SuiteSisterMary · · Score: 2

      This is also, as I recall, what lets you yank out random processors while the system is running, and plug them back in. The problem is that Linux for these puppies doesn't support most of this functionality; there's no way to tell it "Hi, I know you're happily running, but here's an extra 512 megs of RAM for that database process. Ta!" At the moment, this is basically a way to consolidate a rampaging buttload of little servers into one big honking server; one could do much the same effect by buying a really really big x86 box, running one copy of linux, then using that to run (number of processors -1) copies of VMware, set processor affinity, then use those to run images of Linux.

      --
      Vintage computer games and RPG books available. Email me if you're interested.
  82. Solaris free only for smaller systems... by gen11 · · Score: 1

    Solaris is only free on 8 or fewer processors. I don't know what they're charging above that but can only guess that after you've spent a cool half-million on an Enterprise server, the $5-15k for the OS is just a drop in the bucket.

  83. probably you by Ixe · · Score: 1

    ...always safe to trust a penguin... no offense "other guy" but I hate er "strongly dislike" it when ppl bash things I support especially when they just yell rather than supporting their views with facts.

    --
    Sigs pose an operational security risk and help the baddies aggregate data. I guess commenting does too, oops.
  84. What's wrong with that? by Zelet · · Score: 1

    There is nothing wrong with paying for more performance. Plus, what would cost the customer more? Having a monkey come in and flip a switch, or having a new Mainframe made/remade for the extra performance?

    Also, this way the hardware upgrade is basically instantaneous. IBM is just giving customers what they want.

    --
    ...And when they came for me, there was no one left to speak out for me." - Martin Niemoeller (1892-1984)
  85. Amen by Ixe · · Score: 1

    Preach it brother

    --
    Sigs pose an operational security risk and help the baddies aggregate data. I guess commenting does too, oops.
  86. Why Not? by aviator · · Score: 1

    I don't see the problem having Linux running on IBM zSeries hardware (aka MVS, S/390, mainframe, etc.). If it works, it works. And I've never seen anything better than IBM's mainframes for parallel I/O activity on huge scales...

  87. the article makes at least /one/ good point. by Penis_Envy · · Score: 1

    In that it doesn't make sense to spend all that money on a large mainframe ($400k was quoted in article) when you could get inexpensive (and fast) i386 boxes and a load balancer. Server management might be easier with the mainframe, but you could have a couple hundred thousand dollars left over to spend on additional console monkeys. Pardon me if someone's already said this.

  88. Yes, Virginia, there is a Linux standard by yerricde · · Score: 1

    I'm sure you mean "the single standardized Linux distribution."

    No, grandparent referred to the Linux Standard Base.

    To pre-empt the old joke: "All your Linux standard base are belong to the Free Standards Group."

    --
    Will I retire or break 10K?
  89. [OT] is that the new math? by Anonymous Coward · · Score: 0

    $0.025 rounds to $0.3 ?? Wow. That's some serious rounding error. Or perhaps you meant $0.03.

    Personally, I'd just chip in my $0.025 twice, and hand the guy a nickle.

  90. Sun still has no *real* enterprise level hw/sw by mikolas · · Score: 1, Insightful

    Maybe Sun is jealous because they *still* have no real enterprise level HW/SW combination that can do real clustering and partitioning. Just compare the stuff Sun can offer with their competitors (HP and IBM). Sun-cluster has downtime (at least with certain applications) up to several minutes and that is not what I call enterprise level operating environment. With IBM HW you get partitioning and abstraction that will you replace every single component in the system with zero downtime. And what comes to performance, I'd recommend cheap Intel boxen to all customers instead of Sun, they're basically identical when it comes to capabilities and Intel just runs Java much better (just try WebSphere, ATG Dynamo or even Tomcat).

    I know Solaris is something like the standard Unix in the corporate world and I hate AIX and HP/UX, but the HW just is so much better. Now, with the rise of Linux, you get the superior HW with decent Unix-like OS.

    1. Re:Sun still has no *real* enterprise level hw/sw by Drazi100 · · Score: 0

      And what comes to performance, I'd recommend cheap Intel boxen to all customers instead of Sun, they're basically identical when it comes to capabilities and Intel just runs Java much better (just try WebSphere, ATG Dynamo or even Tomcat).

      yeah will see after a few connections which dies like a bitch and it wont be sun

    2. Re:Sun still has no *real* enterprise level hw/sw by mikolas · · Score: 1

      Maybe you should really check out the performance you get with relatively cheap Intel-compatible HW. Basically all a Sun box can offer is fast I/O and you can achieve that with Intel-based server if you have capable RAID controllers and multiple NICs in place. You will have the downtime though, as majority of Intel-based HW is not as hot-pluggable as real enterprise level HW. If you want to have that, check out IBM's OS/390 portfolio (or whatever zSeries that is) or HP's SuperDome. With Sun, you're left with overpriced HW with slow CPUs, no enterprise level uptime and not that good hardware abstraction.

      Just that it runs non-Intel architecture as default does not make it that good. At work we have pretty good uptimes for the Intel boxes running Windows 2000 and especially those running Linux :-) So in my opinion the Unix-hardware-is-better-than-standard-Wintel is just a myth. _Good_ Intel based server HW is just as good as standard Sun stuff, you just don't have to pay the extra for the brand...

      And you really should do those benchmarks, a Sun R440 that costs $$$$$$ just gets beaten when compared with a Compaq's dual CPU server that costs around USD 5000. And this is just something I have personally worked with, maybe you have other kind of experiences.

    3. Re:Sun still has no *real* enterprise level hw/sw by fw3 · · Score: 1
      no kidding!

      The company I used to work for threw $80million! into a SAPR3 system. Andersen Consulting had the contract and they just put the hardware out to lowest bid, no critical eval of capacity to actually achieve better than 99.9% uptime. Sun won the bid.

      Now I'm sure that Solaris *can* do this better but this installation was awful. You could count on a 15 minute outage at least once a month. That meant 4000 employees US, EC, japan etc. unable to access the systems, and it was usually a hardware problem on some Solaris database server.

      Worse, about 2x in a year the whole shebang would go down for a _whole_day_ due to a problem in the cluster. And the (very expensive) full duplicate backup systems never once were able to pickup the load.

      Sorry I didn't think much of sun/Solaris before that, thought much less of 'em after. I make that system's uptime based on experience to have been 99.4%. nowhere near good enough for what we paid.

      --
      Linux is Linux, if One need clarify their dist: <Dist>/GNU Linux
      bsds are of course just BSD
  91. IMHO, this guy has no clue ... by x+mani+x · · Score: 3, Interesting

    Read between the lines, this article is mostly anti-IBM FUD. It was written by Sun, so I'm not exactly surprised.

    And Linux isn't designed to run in a virtual machine; implementation decisions that make sense on PC hardware don't fit well in a virtual machine(4). This is Linux. It's designed for Intel. It's not tuned for the mainframe hardware in which it's running.

    First, let's check what his "(4)" reference points to:

    (4) For example: Filling all available RAM with file buffers is great in a real machine (as it speeds I/O via caching with otherwise-wasted storage), but in a virtual machine doing that is bad (as it inflates the working set of the Linux guest, which is competing for real storage with many other Linuxes-leading to paging/swapping).

    Uhh, I have never seen a VM implementation that did not give a RAM limit. So this guy is basically saying that a memory leak on one of your VM's will take down the entire mainframe. Somehow I doubt IBM's mainframe R&D staff would do this ... unless IBM mainframe R&D is actually a computer camp for children with down syndrome.

    Often the difference in Intel versus mainframe applications makes porting difficult(10)

    (10) Intel uses something known as little endian; a mainframe uses something different. This is significant for certain applications and makes the port difficult.


    I challenge anyone out there to name any significant piece of UNIX software that doesn't have a big-endian port ... uhh basically they don't exist, because many of the commercial UNIX systems out there exist on top of big-endian hardware.

    Just the way he phrased that last bit about endianness convinces me that this guy doesn't have a clue what he's talking about. I can't really know for sure though, since most of the stuff he talks about is beyond me. But, based on those few things he mentions that I'm familiar with, I'd say he's a typical manager who is loosely and incorrectly paraphrasing what some Enterprise developer told him, and decided to make a marketing advantage out of it.

    Read between the lines!!

    1. Re:IMHO, this guy has no clue ... by Anonymous Coward · · Score: 0

      IBM mainframe R&D is actually a computer camp for children with down syndrome ???

      probably not ibm, but you...

      It is a comparison of shared disk cache / per VM disk cache.

  92. Hard to port? Wait a minute... by hansendc · · Score: 1

    Often the difference in Intel versus mainframe applications makes porting difficult(10).
    (10) Intel uses something known as little endian; a mainframe uses something different. This is significant for certain applications and makes the port difficult.


    Wait a minute... Isn't Sparc Big-Endian too?

  93. Why doesn't Sun worry about their own issues? by markj02 · · Score: 2
    IBM is doing well. And I think IBM has smart customers who can figure out whether buying a machine for several million dollars running virtualized Linux makes sense or not.

    Sun should rather worry about their own licensing issues and their own problems with open source. Java's broken community licensing program and Sun's inability to evolve the platform more quickly has basically killed Java for open source applications (that's why Mono is being written around .NET, even though .NET is much less mature and comes from Microsoft). Sun keeps equivocating on Solaris, Linux, and which one is better in their not-so-humble opinion.

    Sun should address their own issues before putting down IBM. I'm sure ten years from now, IBM is still going to be around. I'm not so sure I believe the same thing about Sun.

  94. Sounds like the wrong tools by Colin+Smith · · Score: 2

    The IBM architectures you decided on are exotic compared to the Sun Quad SMP box and don't sound like you benchmarked the appropriate tools, or even like against like.

    Mainframe: I/O throughput, *massive* loads.
    SP: *parallel* clusters, CPU, *fast* network.

    You'd have been better with an S80... Sorry, "pSeries".

    --
    Deleted
  95. honest question by kin_korn_karn · · Score: 1

    call me naive, stupid, whatever.

    other than backwards compatibility and data warehousing, what do you use these mainframe-class boxes for these days? Why run thousands of instances of Linux on one mainframe? the only reason I could think of would be a development environment, but if you're doing that much Linux dev everyone should have their own workstations..

  96. Does a proprietary underlying OS matter? by yerricde · · Score: 1

    Linux on mainframes is still running on top of a properitery VM. This kills the whole Free Software argument for using Linux.

    Likewise, Linux on a PC farm is still running on top of a "properitery" BIOS (unless you're using something odd like LinuxBIOS). This kills the whole Free Software argument for using Linux.

    But does it matter?

    Of course, why would a company endorse another company's product?

    When it sells products that are designed to interoperate with the other company's product perhaps? (That is, not in this case.)

    --
    Will I retire or break 10K?
  97. This might be an interesting story if by electroniceric · · Score: 1

    somebody put Sun's spin on Linux against IBM's against ESR's against Microsoft's. Pick apart the so-called technical arguments and see how the Linux-in-the-enterprise battle is fought.

    By itself, this is just the bulleted text on the back of the box the Sun comes in.

  98. flaw? by DeadPrez · · Score: 1

    Applications that run on Linux for Intel need to be recompiled and recertified for each new platform; thus the application portfolio to run Linux on a mainframe is small.

    Then later (as a reason Sun/Solaris is better)...

    Linux compatibility features of Solaris

    Perhaps this is just a strawman but just because IBM certifies programs and Sun emulates (but not certify) doesn't mean Sun's solution is more l33t.

    In fact, in production environment would you rather run programs certified by IBM or trust Solaris to emulate linux well enough to run the same programs. No-brainer...

  99. What do slashdoter's think? by autopr0n · · Score: 2

    What do the folks on Slashdot think?

    Well, I can't speak for others, but I think sun sells a competative UNIX on mainframe solution.

    --
    autopr0n is like, down and stuff.
  100. What about this?? by bogie · · Score: 1

    "Suited for Web delivery at the edge of the network, Linux servers offer an alternative to proprietary and closed environments such as Microsoft Windows"

    So linux is only good for light web serving duty?

    "implementation decisions that make sense on PC hardware "

    Read between the lines linux is for low end "pc class hardware"

    "Applications that run on Linux for Intel need to be recompiled and recertified for each new platform; thus the application portfolio to run Linux on a mainframe is small("

    here we see the time honored no apps FUD.

    "Why put an open operating system such as Linux on a closed proprietary mainframe?"

    cough....did he actually say that...cough

    Sun loves to dam linux with faint praise. The only reason they even have a linux "strategy" is because of fear of losing market share.

    Me thinks the dot in .com is fast becoming a dud.

    --
    If you wanna get rich, you know that payback is a bitch
  101. Not a good idea? Maybe, but... by frozenray · · Score: 5, Insightful

    ...there are other factors than performance, indulge me for a moment:

    I work at a big corporation which relied on IBM mainframes for its whole business for almost 30 years until the PC and the high-end Unix servers shook up the landscape for good. I'm from the PC (IT) camp, which has been separate from the Big Iron (DP) guys in the organization since the early days.

    DP, once very powerful, has lost a great deal of influence in the 90s, although they still run most of the mission-critical stuff, and the main reason for this were the high-end Unix servers, most of them Sun boxen running Oracle. Believe me, there's no love lost between those two fractions in our company.

    Our mainframe guys see Linux as an opportunity to get better integration with the IT world, which was abysmal until now (3270 terminal windows, IMS/DB, TSO/ISPF and such horrors) and to better position themselves against the Sun/Oracle camp which is after their budgets and their butts. Today, we have Linux happily running on our mainframes (still in an experimental phase, not in production), serving up http and Samba shares without a hiccup.

    If we're talking about bringing Linux into the large corporations, the crucial influence of IBM cannot be overestimated. We were a died-in-the wool IBM shop (S/390, Token Ring, 3270PC, OS/2, S/36, AS/400, the whole enchilada) and successfully trusted our business to IBM for 30 years (paid through our nose for it, too, I might add). IBM has lots of credibility and trust, so if they say Linux is cool, our CTO listens. Microsoft, on the other hand, is viewed with some "new kid on the block" suspicion. Our management doesn't like downtime and security breaches, and the memory of the ILOVEYOU aftermath is still very vivid, for example. Plus, we migrated to NT4 late (about 28'000 systems, ended September 99) and now Microsoft is practically forcing us into another expensive upgrade cycle sooner than we wanted and with IT budgets cut short on account of the less-than-stellar economy because NT4 support is withdrawn in 2003.

    We thus have the following situation: IT and DP are up against the Unix enterprise server guys, all this with the backing of IBM. The astronomically high cost of Sun/Oracle solutions is being questioned more and more, and technologically viable low-end solutions (x86 multiprocessor servers, Linux) begin to rattle the foundations from below.

    I don't want to make bold predictions here, but if I were Sun, I'd be worried. To me, it looks like interesting times are ahead.

    --
    "There are already a million monkeys on a million typewriters, and Usenet is NOTHING like Shakespeare." - Blair Houghton
    1. Re:Not a good idea? Maybe, but... by msouth · · Score: 2
      I don't want to make bold predictions here, but if I were Sun, I'd be worried.


      And, if I were Sun, and I were worried, I would probably send up a FUD balloon and see which way the community reaction wind is blowing.

      --
      Liberty uber alles.
  102. VM removes the need for load balancing software by jms · · Score: 4, Insightful

    Linux on the mainframe can't respond to the workload demands of Web serving with high utilization--something IBM touted at the time of its z800 announcement. Horizontally scaled Linux farms are designed to handle unpredictable demand with above average peak loads. As demand rises, a load balancer distributes the traffic evenly across servers, which increases utilization. Because design capacity needs to handle peak demand, server farms often have a low utilization.

    If you have a VM system with two virtual machines, and one of them is nearly idle, and the other virtual machine is very busy, VM will automatically take resources away from the less busy machine and devote it to the more busy machine.

    This means that you don't need load-balancing software. VM is the load-balancing software.

    Given the relatively low cost of hardware, some organizations find this trade-off acceptable to ensure appropriate service levels. Contrary to what many believe, consolidating a Linux farm into multiple images on a mainframe would not change the demand pattern. Although z/VM can start and stop Linux images, it cannot dynamically add resources to match demand.

    Of course it can! The VM kernel will parcel out memory and CPU on demand.

    As a result, a mainframe would need to size for peak demand just as the Linux farm would;

    All computer systems need to size for peak demand. The difference is that with a mainframe, you can size one machine for the peak demand of the busiest of a large number of virtual machines, and get rid of the overhead caused by the load-balancing software, because you don't need it anymore.

    high utilization is a myth.

    VM systems can utilize 90-95% of the native computer resources. The overhead on a VM system is very, very small.

  103. It's FUDerific!! by borgheron · · Score: 1

    This is nothing but marketing BS and FUD. Sun is loosing a great deal of business recently due to the introduction of Linux into the server market.

    GJC

    --
    Gregory Casamento
    ## Chief Maintainer for GNUstep
    1. Re:It's FUDerific!! by LoseNotLooseGuy · · Score: 2, Informative

      Sun is loosing a great deal of business recently due to the introduction of Linux into the server market.

      I find that unlikely. I think Sun's stockholders would be very upset if Sun were actually "letting loose or releasing" business. However, it could be argued that Sun is failing to retain business due to Linux deployment. The word you were looking for is losing.

      Congratulations! You have been participant #35 in my campaign to rid Slashdot of this error.

      --
      Proudly correcting Slashdot's most irritating linguistic error since 2002.
    2. Re:It's FUDerific!! by craig42 · · Score: 1

      I am normally a lurker, but I must post to thank LoseNotLooseGuy for his efforts to rid Slashdot of this great annoyance!

    3. Re:It's FUDerific!! by LoseNotLooseGuy · · Score: 1

      craig42, your support is gratefully noted. You have been counted amongst the Supporters of the Cause.

      --
      Proudly correcting Slashdot's most irritating linguistic error since 2002.
  104. mostly waffle by Anonymous Coward · · Score: 0
    The very title of this dude ("Chief Competitive Officer") should be a Big Clue.

    Firstly, the mainframe-with-lots-of-virtOSes is a much neater solution than chroot for making lots of boxes for particular purposes while using very little hardware. Chroot is fiddly to set up and maintain, especially if you have minimalist chroot environments, as is normal.

    The "designed for Intel" argument isn't totally without merit, but mostly means that most of the really fine detail per-architecture optimisation effort goes to Intel platforms, not that the OS itself has much bias. After all, Linux runs on a scarey number of platforms. If there were strong bias this wouldn't be the case.

    The virtual machine argument is similarly weak: the IBM hardware and supporting OS is designed for virtual machine presentation, and can be expected to be fairly good at it. It's not like the Linux instances are running in an emulator, with interpreter-type overheads.

    Looks like FUD to me. It fairly shouts it at me!

    Sun's rep is built on solid, standard, hardware and a solid, scalable, enterprise featured OS. Look at the way cool things you can do to an E10K machine: hot swap damn near anything, run multiple Solaris instances and move CPUs etc between them, on the fly, etc. This doesn't mean the IBM Way is silly. A different set of benefits.

  105. In Sun's defense by RoosterT · · Score: 3, Insightful

    Naturally this article is going to be met with some skepticism because it appears to be a self-serving marketing piece from Sun. This is unfortunate because the author makes several good points. I think it is important to note that this is more of an attack on mainframes and VM architecture than it is against Linux. It really does not make much sense to run 10, 100 or 10,000 copies of Linux on one super-duper computer. Sure it's neat, but we need to remember that computers are supposed to do useful things. What a collosal waste of cpu and memory to have 10,000 operating systems when the right one (yes, 1) would do the job just fine. Add to that the inherent performance hit from running on a "virtual machine" and it makes even less sense. I actually tend to think that IBM's use of Linux is more self-serving than Sun's attack. It's just an attempt to sell more expensive proprietary hardware by capitalizing on the intellectual generosity of others.

    1. Re:In Sun's defense by chris_7d0h · · Score: 1
      Of course IBM aims to sell more "expensive" hardware with the help of Linux! IBM is a company and does not aim to do stuff which make them loose money in the end.
      Sure it's neat, but we need to remember that computers are supposed to do useful things
      And how does this relate to running progams on a specific architecture?
      What a collosal waste of cpu and memory to have 10,000 operating systems when the right one (yes, 1) would do the job just fine
      I take it you've never been in a situation where you needed the ability do partition resources and access for specific purposes, people or customers? As for the waste of resources. Do you know anything about how an S/390 handles resources?
      Add to that the inherent performance hit from running on a "virtual machine" and it makes even less sense
      &ltsarcasm&gt Ah, yes, and what an "incredible" performance hit it is...&lt/sarcasm&gt
      I actually tend to think that IBM's use of Linux is more self-serving than Sun's attack. It's just an attempt to sell more expensive proprietary hardware by capitalizing on the intellectual generosity of others.
      Right. Thank god that SUN isn't using proprietary hardware, because that would make SUN just as bad.
      --
      In a society that believes in nothing, fear becomes the only agenda ~ Bill Durodié
    2. Re:In Sun's defense by RoosterT · · Score: 1
      And how does this relate to running progams on a specific architecture?

      It means use the right tool for the right job (a mind boggling idea for many Linux zealots). If the only tool you have is a hammer, everything looks like a nail.

      I take it you've never been in a situation where you needed the ability do partition resources and access for specific purposes, people or customers?

      Gee...and i here i thought it was the operating system's job to manage system resources...

      Ah, yes, and what an "incredible" performance hit it is...

      My bad, I guess that's why C++ is dead and everyone uses Java now...

      Thank god that SUN isn't using proprietary hardware, because that would make SUN just as bad

      For the record, I never claimed Sun was somehow better, just that IBM does not exactly embody the open source ideology of Linux

  106. what good quality stuff has come from U.S by Drazi100 · · Score: 0

    micorsoft appears to be another bloated amercan company that cant make reliabele product

    microsoft is like GM , it dominated once, but it always made unreliable crap.

    the japs should make their own OS. maybe they will finally get rid of the kludge knonw as
    wintel once and for all

    1. Re:what good quality stuff has come from U.S by LarryWest42 · · Score: 1

      That's just silly.

      Unix, C++, Xerox Star, Ethernet, TCP/IP and friends, MacIntosh, Quicken, Java, Perl, C#, ... lots of quality stuff has come from the U.S., usually with strong contribution from foreigners (Bjarne Stroustrup, etc.).

      That's why it works: we in the U.S. don't really care where it's invented or by whom (apart from English language bias and ignoring corporate NIH as a different problem).

      If you think "Japan" can make its own OS, you should review the disastrous history of the "Fifth Generation" plan (I think that's the name) from the late 1980s.

      Certainly Japanese companies, universities, or individuals would be capable of creating a solid OS. But they'd be wise not to exclude the knowledge and experience of people in other countries.

      And unless you're going to do something fundamentally new [perhaps non-von Neumann systems], I don't think writing a new OS is really the best area to have your smartest people working on at this point.

  107. Re:Linux at any cost by Anonymous Coward · · Score: 0

    Au contraire, mon frere. If you read the article and sampled the footnotes you'd realize that the instances of Linux on these IBM mainframes are entirely virtual and lose the strangths of Linux running natively on the hardware.

    The Linux PHILOSOPHY is great, but the implementation is (relatively) poor: native apps are much faster in this case, and more importantly, less dependant on a single point of failure. Virtual machines are neato when used intelligently, but I think the author is making a very strong point regarding IBM's over-dependance on Linux *at all costs* as opposed to choosing the best tool for the job.

  108. Yeah, right - Linux not ready for Mainframes by Colin+Smith · · Score: 2

    Of course, IBM on the other hand, took linux seriously and now have a profitable product that is being very well received in the market and which *is* saving businesses lots of money.

    Sun doesn't have anything quite like it.

    --
    Deleted
  109. Sun just wishes.... by maxpublic · · Score: 2

    It seems lately with Sun floundering about the market and pulling silly moves re Java that all this really points to is that Sun really, really, really wants to be Microsoft.

    And they aren't. And they're pissed about it.

    Imagine for a moment that Sun has the hugely dominant market share in server revenues that it wished it had, and that cross-platform Java programming and their version of .Net rules the world. Do you honestly think that this company would be any more ethical than Microsoft?

    Given the way they've been acting, I think that Sun - if it had the opportunity - might even turn out to be *less* ethical. Much of what they've done lately reminds me of a five-year-old screaming "my toys are better than your toys! And they're my toys! And you can't touch them unless I say so, and only if you'll play with them the way I want you to!"

    I don't trust IBM any more than I do Sun when it comes to their motivations re Linux, but at least IBM has some class....

    Max

    --
    My god carries a hammer. Your god died nailed to a tree. Any questions?
  110. Linux Running on IBM Solaris Running on Sun by Anonymous Coward · · Score: 0

    Everyone seems to have forgotten the quietly mentioned announcement that was inserted in the big announcement about Sun selling Intel-powered servers running Linux. Namely, Sun fully intends to create its own distribution of Linux and to bypass RedHat and Suse.

    Sun fully intends to hijack Linux, which is the core of the open-source movement. Seize the OS. Then, all the applications are at your mercy. (This whole thing smacks of Microsoft and its strategy for Windows).

    Note that IBM is working fully with RedHat and Suse. IBM is not creating its own distribution of Linux.

  111. Can we slashdot a mainframe? by Colin+Smith · · Score: 2

    Hmm? Can we, can we?

    What d'ya think? Anyone want to post odds?

    IBM are giving out VMs on some of their mainframes.

    --
    Deleted
  112. RAS (was I disagree) by dunstan · · Score: 2

    Sun's big pitch is RAS (Reliability, Availability and Serviceability), and they have got it absolutely spot on with the new F15K. The dynamic system domains on the 10K and 15K (and the next size down Serengeti machines) allow multiple Solaris instances to run on a single frame, and for processors, memory and I/O to be put into and taken out of a running Solaris instance. But unlike VM, the domains run in partitioned sets of hardware, so that (for instance) any CPU fault will only bring down a single instance, whereas with VM style logical partitioning you could lose a whole lot of domains with a single hardware fault.

    No, this article is about the fact that "Linux Mainframe" sounds like the benefits of the Linux API with the RAS of a mainframe, whereas in reality you are getting something which is not as resilient as a E10K or F15K, and the Linux kernel won't linearly scale out to 72 CPUs and 500GB of memory as a Solaris kernel will.

    GNU/Linux on mainframe has its place, all Sun are trying to do is point out the IBM might get a bit carried away by what they achieve, and try to pitch it into situations where a different system might be better.

    Dunstan

    --
    The last scintilla of doubt just rode out of town
    1. Re:RAS (was I disagree) by Raven667 · · Score: 2
      But unlike VM, the domains run in partitioned sets of hardware, so that (for instance) any CPU fault will only bring down a single instance, whereas with VM style logical partitioning you could lose a whole lot of domains with a single hardware fault.

      I believe that you are mistaken. On an IBM system if a processor smokes all the tasks that wer e running on it will be automatically moved to a redundant spare and execution will resume from the last known good state. The Linux image would continute processing normally and would be unaware that anything at all had happened. I also believe that the machine would generate an exception alert for the local admins, as well as call IBM Global Services directly and report the problem so that a tech can already have been dispatched by the time the customer calls reporting a problem.

      All the documentation that I have read indicates that IBM VM is significantly more advanced than Sun partitions. All the whining from Sun won't change that anytime soon, only several years of hard engineering will close the gap.

      Linux on a mainframe can be great if you have a computing task that matches what mainframes are good at. It can also be good if you have an existing mainframe and wish to get more use out of it.

      --
      -- Remember: Wherever you go, there you are!
  113. Linux runs faster than Solar? by Maddog_Delphi97 · · Score: 1

    I've heard that Linux runs faster on Sparc hardware than Solaris... is there any confirmation of this?

  114. If it's not good, then fix it! You have the soruce by trenton · · Score: 2
    Shahin Khan says:
    This is Linux. It's designed for Intel. It's not tuned for the mainframe hardware in which it's running.
    The Open Source Software community says:
    This is Linux. It's designed to be modified. You can tune it for whatever hardware you want, since you have the source.
    There's no way you could tune Soliaris for anything. It's designed to run only on Sun hardware, which is great if you're Sun.
    --
    Too big to fail? Does that make me to small to succeed?
  115. Well... by LiquidPC · · Score: 1

    I see all these people flaming sun for "bashing" linux, i put it in quotes because I don't see it in the article anywhere _bashing_ linux, however, I doubt 90% of these people have _ever_ used a mainframe and are just sticking up for linux because they use it, some would call them zealots. In my opinion, I feel sun has made some very interesting and well-founded points that Linux wasn't created for the mainframe, it was created for x86 and not to run in virtual machine, etc. On the other hand, it's _your_ mainframe, and you can put whatever you feel like on it.

  116. Context is Worth 80 IQ Points by Anonymous Coward · · Score: 0

    What serves Mr. Khan so well is his agenda (getting people to do things the Sun Way.) What does not serve him well is his implied ignorance of the kind of environment where Linux on a z-series might actually make sense.

    That is, in mainframe shops, NOT where he supposes it will be introduced. He paints a relentless picture that a shop would purchase a brannu z-series JUST TO RUN LINUX. That might be true in a mainframe shop, with trained folks who are used to the environment.

    A UNIX shop would probably not ever find itself in this position unless they have a particularly thick headed CIO who comes from a mainframe shop and doesn't understand the UNIX Way.

    So the thesis of Mr. Khan's paper is bogus - painting a picture of painful adoption by an entrenched UNIX shop with architectures built to suit, and thus this paper is simply bogus as well. I would expect better of someone with such a lofty title. Perhaps he's embarrassed of this paper. But he's probably just another willing lapdog in the kennel of McNealy (whatever you say, boss!)

    Appropriate technology applied appropriately, Mr. Khan. That's the way to gain credibility in IT shops, not trying to presuppose a scenario that's self serving, not to mention silly.

  117. A few AC comments. by Anonymous Coward · · Score: 0

    He says "Linux is designed for Intel" That is incorrect, Linux is designed as a mostly portable operating system that currently runs on several architectures INCLUDING Sun, IBM RS/6000, Alpha, Macintosh, and others.

    He complains about recompiling your software to run on the mainframe. Well big deal, if you are running the same libs and same compiler with a same kernel then the compile will be relatively painless! Its like compiling Solaris Sparc software to Solaris Intel. Sun claims its easy.
    And now I hear that they are going to sell Intel machines running Linux?!
    Well thats going to be a bit trickier compiling Solaris Sparc programs on Intel Linux than Intel Linux on S/390 Linux.

  118. Calling the kettle black by blang · · Score: 2

    Putting linux on a E10K is just as dumb as putting it on a mainframe.

    Big irone are priced at a premium.
    For one thing, resources are shared among CPUS, so you can't scale linearly, giving you less bang for the bux, the bigger the box.

    What is worse, is that the price tends to cale almost exponentially.

    For example a beefed upentry level server with the same performance as a half full mid-range/enterprise server typically costs only the third of the bigger box.

    Why would anyone want to buy such expensive boxes to run linux? The only reason I see is that they think they can save money by having fewer linux admins. Me, I think you can buy a lot of linux admin payroll for the 5-10$ million premium you pay for a big iron. I also think that an operation with some big iron tends to require a higher head count than the same computing power on commodity servers.

    The only places I see a market for the big iron, is where they're just ading software and capabilities in an operation where they already have big iron. I.E IBMS's and suns goal with this, is to stop existing customers from throwing out their existing machines.

    --
    -- Another senseless waste of fine bytes.
    1. Re:Calling the kettle black by mikolas · · Score: 1

      Why would anyone want to buy such expensive boxes to run linux? The only reason I see is that they think they can save money by having fewer linux admins. Me, I think you can buy a lot of linux admin payroll for the 5-10$ million premium you pay for a big iron. I also think that an operation with some big iron tends to require a higher head count than the same computing power on commodity servers.


      Do you have any kind of understanding how much does the physical space in a hosting facility cost? Not to mention about the setup cost. With one *huge* box running several Linux-instances you can easily save some serious money if you are a corporate user with a need for some hardcore hosting (a datacenter or such). Linux on IBM mainframe is not aimed at casual users, it's aimed at the big corporations... Just call HP or IBM Global Services and ask how much does the hosting cost. Of course, you can get cheapo hosting somewhere, but we're talking about corporate users and big investments here. With low-end hardware you'll end up adding more boxen based on current need and pay a huge setup costs per box, with IBM's mainframe you'll just add another Linux partition. Simple as that.

  119. Linux not just for intel by mfos.org · · Score: 1

    Did anybody else catch that bit about linux being only for Intel?

    A quick ls /usr/src/linux/arch | wc counts 15 different architectures, not to mention all of the other architectures not in the main kernel.

  120. Some Inaccuracies Here... by Anonymous Coward · · Score: 1, Insightful
    Where do I start...
    • Linux on the mainframe is actually hosted by another proprietary operating system, z/VM.
      Not necessarily, you can boot Linux as a stand- alone operating system. It's just more convenient to develop and implement under VM in many shops.
    • The optimized operating system for IBM mainframes is z/OS, formerly called MVS(2). Compared to z/OS, z/VM is a niche operating system with virtual machine (VM) support for new hardware features added late or often not at all(3).
      There is a grain of truth in this. Some printers, for example, were supported late in VM. I think the same is true of some tape libraries. However, it's a real strong niche, and usually this is not an issue.
    • Linux on the mainframe is complicated; this isn't Linux running on a two-way Intel server. Despite IBM's claims of easy management(5), customers still need a special machine room and specially trained staff for both z/VM and Linux.
      Linux on anything can be complicated. If it follows the historical path of IBM products, it will get both (a) more complicated as IBM adds options and (b) simpler to administer as they clean up the packaging. But the real problem here is the specially trained staff...you need a trained staff no matter where you are running Linux...or any operating system. The same is true of Solaris or Windows servers.
    • Finding mainframe staffing is an obstacle in many organizations(6); combining mainframe and Linux staffing further complicates the matter.
      This is a good point. Mainframe skills are becoming rarer, older, and grayer. But...once you have the VM systems programmer, the rest of the Linux staff can all be new hires trained for IBM Linux. However, finding good Solaris skills can also be a challenge.
    • Running multiple Linux images still requires administration that needs to grow with the number of images being run.
      No. The strength of IBM is scalability. The strength of IBM customers is organization and procedures - you might see it as bureaucracy, but it has gotten the job done for two generations. And, if I am running thousands of Linux virtual machines, I will be able to automate the management of these with existing MVS mainframe tools.
    • ...there is little incremental RAS benefit. Although IBM claims "zSeries servers inherit the legendary IBM S/390 strengths in the areas of fault avoidance and tolerance, recovery from failures, and concurrent maintenance and repair for "always-on" availability"(7). We don't believe this to be true for zSeries servers running Linux. The "legendary" IBM S/390 strengths IBM references are the result of decades of development work on IBM's flagship mainframe operating system, known today as z/OS. The fault recovery features of z/OS are not found in Linux. z/VM does have some fault recovery features, but it is not nearly as resilient as z/OS. For example, z/VM cannot take advantage of Parallel Sysplex clustering, and VM hypervisor is an added single point of failure(8).
      There is a lot of machine check recovery processing in MVS. However, over the past decade a lot of that has moved into the hardware itself, and into the Hardware Management Console (HMC). So, this was certainly true in 1990, less so today. Ditto the hypervisor comment IMHO.
    • Consolidation without application availability just can't happen...
      Unless you are writing the applications yourself, as many potential IBM Linux customers are.
    • Often the difference in Intel versus mainframe applications makes porting difficult(10).
      Little-endian vs. big-endian. Hits Solaris too.
    • The economics just don't work. IBM claims it is financially justifiable to consolidate as few as 20 Linux servers on a z800(11). With an estimated starting price of $400,000 for a z800(12) with a single CPU engine enabled, that claim seems exaggerated compared to Linux servers that hover in the $1,000 to $2,000 range. Sun's rack-optimized 1U form factor servers start at list prices under $1,000.
      Just looking at hardware and/or software costs isn't the right analysis for most of these customers. You also have to factor in the cost of support staff; comparing z/VM and Linux to a corresponding Windows NT server farm becomes more competitive as the size grows. And factoring in the cost of downtime, which is the real driver for the very large companies who benefit the most from this combination, usually makes the hardware costs insignificant.
    • Customers such as Nielsen Media Research, A.B. Watley, Cognigen Corp, and Littlewoods have chosen Sun and realized up to $1.5 million in annual savings
      Neilsen is a household name, haven't heard of the others, but...these are hardly huge DP shops. The real benefit of z/VM and Linux is going to be in the Fortune 100 manufacturers, insurance companies, banks, and stock brokerages whose IT budgets approximate $1.5 million per week. Or day.
    • [out of order] This is Linux. It's designed for Intel. It's not tuned for the mainframe hardware in which it's running.
      This is Linux. It's designed to be open. It's tunable for any hardware.
    Don't get me wrong...I love Sun, have an Ultra 5. But a lot of this is obfuscation. There is a lot of value in Linux on the mainframe for the huge organizations which are, in the main, already IBM customers.
    1. Re:Some Inaccuracies Here... by dgavin · · Score: 1

      This SUN whitepaper sounds familiar - I used to read the type of FUD from IBM when I worked for Amdahl supporting UTS (SYSV UNIX) on their multi-engine, multiple virtual image mainframes. We had a lot of similar issues: lack of apps, lack of technical staff that understood both worlds, etc. Most of the customers (RBOCS mostly) were writing their own apps or getting them from BellCore, the others were using them as massive DB servers or Office replacemnts for email and scheduling. In it's time, it was a good solution and despite all of IBM's FUD, the customers were generally happy with the platform.
      I guess if you wait long enoough, everything comes around again in a slightly different guise 8-)

  121. Re:If it's not good, then fix it! You have the sor by korgull · · Score: 1

    Why tune if another thng runs better ?

    No bloody manager would decide to do that.
    And he would be right. it'll cost him a lot of money to modify the source and money is all they care about (mostly).

    I do agree that they can change the source and in the long run they might decide to do that. But, it's not a short term option.

  122. Sun is the one who loses here. by Anonymous Coward · · Score: 0

    In our company we have pretty much gotten rid of Sun hardware & software and replaced those with mainframes running Linux. We've been very satisfied. Just like Telia - we do not need 70 Sun servers if we can replace those with one single IBM Linux mainframe which is easy to manage and which can handle the job. Business is business. I can understand if Sun gets mad about this, but I couldn't care less. Our Oracle server runs on Linux platform (it used to run on Solaris) and we have not had any problems with it ever since we switched from Solaris to Linux.

  123. Flame me, I'm already burning. by Anonymous Coward · · Score: 0

    I don't get it. I don't get it at all.
    First of all, don't start slating me as an 'enemy'. I think Linux is great. I've been running it at home for 5 years (yes, yes .. I /know/ that there are people out there that have been using linux since version -0.004 .. this isn't a p155ing competition you know ...).

    Having said that, I've had experience with a wide range of computer systems ranging from embedded with a couple of K ram to larger (look, it's /still/ not a p1ssing comp.) systems with gigs. I just really fail to see why people expect linux to scale both ways. Desktop users have very different needs to embedded users, as do mainframe users. I see so many posts that are automatically hostile to the "maybe Linux isn't the best choice" point of view ... did you ever acutally think before you but your postings in gear?

    Why do people have such a problem believing that one shoe can't fit all? I'm not saying that Linux is cr|p on mainframes, but I am speaking out against the knee-jerk reaction which says "if it's Linux it can't be wrong!".

    This isn't flame bait, it's a realistic view. If you can't hack it, don't hack it and don't inflict your unfounded views on others.

    1. Re:Flame me, I'm already burning. by kberg108 · · Score: 0

      stay on the subject your confusing us :)

      --
      I like things that are sweet and not things that are lame. --
  124. Cost isnt as big issue on a mainframe perhaps? by miffo.swe · · Score: 1

    But as linux is pretty stable in comparison to Windows it cant be that bad to run it on a mainframe. Sun has to jump onto the linux train to be able to compete with IBM but they dont seem happy about it.

    I wonder why, is it because they are behind in emnrading linux?

    Tossed the windows partition out today and i feel free. Never ever going back!

    --
    HTTP/1.1 400
  125. Its a pity the article contains so much crap.... by maroberts · · Score: 1

    ..'cos I think if you filtered it and thoyght about it you could make quite a good case as to why Linux on a mainframe such as this is probably not the best idea in the world at the moment and has some way to go before it becomes a really good OS for such a machine.

    Instead the author seems to have got rooted into the idea that Linux is only for Intel processors really and can't handle things like big/little endian swaps.

    --

    Donte Alistair Anderson Roberts - hi son!
    Karma: Chameleon

  126. jealous by Anonymous Coward · · Score: 0

    Sun is just jealous because solaris sucks!!! And I base this on my actual experience for using both solaris and linux. And I can say from my experience solaris sucks!!! I figure this is just SUN fud.

  127. Open Sourcers offering free rent in Condo Complex. by Anonymous Coward · · Score: 0

    Open sorcers have annouced that money is no longer needed in the open sourced world and are giving all material belongings away for free. Free rent on all open sourcers real property is the first evidence that these money haters are putting there dollars where there mouths are! Also up for grabs are their bank accounts. We have finally reaced either the twilght zone or the outer limits. The commodity know as cold hard cash has finally become meaningless. Everything is free!

    Open Sourcers will soon announce the code for digital molecular reoraganization allowing all human beings to have any food or beverage or any materila thing they they reqiure therefore negating the environmental collapse our overpopulated world is racing toward.

    Long Live Linux!!!

  128. Considering all the dumb things Sun has done... by restive · · Score: 1

    they probably have/had the potential to be really great, but then have to make changes or statements along the way that leaves people shaking their heads.

    I'm admittedly a Sun bigot of sorts, but articles like this are pretty uncalled for and make them look like schoolyard brats.

    Grow up Sun, you guys have great hardware and a great OS (in my opinion). Stop being stupid in your attempts to get ahead...

    1. Re:Considering all the dumb things Sun has done... by kberg108 · · Score: 0

      I'm affraid I disagree. I am no expert on hardware or OS's by any means but in the last six years I have spent developing software I have seen more problems with the so called stable system know as the SPARC platform. I have seen more SUN products shipped back to the manufacturer than any intel box. From what I understand SUn used to make rock solid servers and a rock solid OS. What I have seen in the past few years from SUN is exactly as you put it a schoolyard bully who looks for any chance to sue any possible competitor. I think that this practice of legal search and destroy they have taken as seriosly effected thier product line and the quality of work they produce. All you have to do is look at the turn over rate in thier engineering department. Any rate I hope they grow up and start treating the open source community with a little more respect and start focusing on creating new technology not just sueing as many so called pattent infringers as they can. I mean really the only thing SUN has going for it is java and they can't even manage that as a real product. Face it SUN is washed up if it wasn't for the few large enterprises that they have been in bed with for years they would be chapter 11 right now. In other words SUN GROW UP you fucking babies. Get back to your engineering and fire your legal depertment becuase there is more to this industry than winning the next pissing match with M$.

      --
      I like things that are sweet and not things that are lame. --
  129. It's mostly FUD by uweber · · Score: 1

    Linux on mainframes kills the serverfarm, be it Solaris/Sparc, Linux/x86 or WinNT/x86 it is a simple calculation of how many people you need to babysit hundreds of machines, the space you need to put em and cooling requirements in comparison to a single zSeries mainframe.
    There was a recent study that stated that you can reduce cost for dayly operations in your datacenter by up to 90% simply because one mainframe uses a lot less power than 500 Solaris/Sparc servers and doesn't require such a large air conditioning system (after all energy costs money, too). Of course this also means that most things that are beeing moved over to the mainframe used to run on some comercial unix and not, like most of us would prefer, on WinNT which makes it clear that Sun is not happy about that.
    And while the Linux S/390 combination might not be completely on par with Solaris/Sparc it sure is not that far of so I guess Sun is extremely worried.

    --
    --Ulrich
    On no accounts allow a Vogon to read poetry at you
    1. Re:It's mostly FUD by kberg108 · · Score: 0

      If you really want cheap buy a full rack of RLX Tech servers running the crusoe proccessor slash power and maintenence by 75% weee go crusoe :)

      --
      I like things that are sweet and not things that are lame. --
  130. Sun is in a rough position by harlows_monkeys · · Score: 1
    IBM has them beat on the big server end.

    Intel has them beat on the workstation end, slightly in absolute performance, and massively in price/performance ratio.

    Microsoft is about to give Java a major ass kicking with .NET.

    They aren't in danger of going under or anything like that, but they've got attacks coming in on all fronts, so should be getting a bit nervous.

  131. Re:Open Sourcers offering free rent in Condo Compl by Anonymous Coward · · Score: 0
    That's got to be the sorriest troll I've ever seen. I hope it didn't take you long to write that.

    ~~~

  132. Partitioning != VM by ansible · · Score: 3, Informative

    System partitioning isn't the same as IBM's VM technology.

    With your E15K, you're dividing up processors and memory between various partitions, each running an instance of the OS.

    IBM's running multiple OSs as virtual machines on the same system.

    With Sun, if partition A is really busy, and partition B is idle, you can't make use of those idle processors unless you re-allocate your partitions.

    With IBM, the processes in VM1 can use all the processors of the mainframe, unless VM2 also needs processing time.

    With either one, if one OS instance crashes, it shouldn't affect the other instances.

    1. Re:Partitioning != VM by charon.de · · Score: 1

      Just wait until the backplane crashes, fix it and it reboots with POST level xy, wow it may take a few hours until the system is up again...

      Michael

  133. Only Partly True. See Gartner Report below by bstadil · · Score: 2

    Published Today by Gartner. Quote

    IBM surged in server market share in 2001, gaining at the expense of Sun Microsystems, Hewlett-Packard and Compaq Computer as the overall market shrank, new figures show. IBM's server revenue decreased from $13.9 billion in 2000 to $13.6 billion in 2001, according to figures from market researcher Gartner released Friday. But because the overall market shrunk faster--down 15 percent, from $55.6 billion to $47 billion--IBM actually gained share. Big Blue cemented its first-place spot in the worldwide market, increasing its share from 25 percent to 29 percent. The increase mirrors similar changes in the North American server market. IBM has benefited from resurgent sales of its old-guard mainframe line, spurred in part by the new ability to run the Linux operating system. But demand for servers in general dried up, with companies worried about the recession and overcapacity left over from the Internet spending spree. Worldwide, second-place Sun dropped 2 percent, to 15.4 percent, while third-place Compaq dropped 0.9 percent, to 13.9 percent, and fourth-place HP dropped 0.1 percent, to 12.8 percent. Fifth-place Dell Computer eked out a 0.1 percent gain to 6.4 percent. In the key Unix server market--a sweet spot with a good balance of server power and price--IBM also gained, increasing share 2.3 percent, to 20.3 percent, with sales of $4.2 billion. The Unix server market dropped 18.7 percent, from $25.3 billion in 2001 to $20.6 billion in 2000, Gartner said. The Unix server market is the biggest single segment of the server market, accounting for 44 percent of total sales. But IBM's gain wasn't enough to topple No. 1 Sun, whose share shrank 3.1 percent, to 35.2 percent, with sales of $7.3 billion. And No. 2 HP, after a concerted effort to stanch losses, rose 1 percent, to 20.5 percent, with sales of $4.6 billion. Dell gained the most in the Intel server market, increasing 0.5 percent, to 17.6 percent, with $3.1 billion in sales. No. 1 Intel server seller Compaq lost 1.6 percent share, dropping to 26.3 percent with $4.5 billion in sales. The overall Intel server market dropped 16.2 percent, from $20.5 billion to $17.2 billion, Gartner said.

    --
    Help fight continental drift.
  134. Hmm this IS an attack by Anonymous Coward · · Score: 0

    not that they might not have some technical backup this clearly is an attack. I'd be curious to know if they'd say the same about *BSD.

    Like someone said on another thread yesterday or day before: Sun just can't let go

    That's so true.. and they have this jewel called Java but they'll probably hold on to that until .NET/#C has per default replaced that... it's really sad

  135. I'm impressed by Watcher · · Score: 2

    This is Linux. It's designed for Intel. It's not tuned for the mainframe hardware in which it's running.

    Wow. In one whole sentence this fellow managed to slap the face of just about every kernel developer since the 1.0 days. A vast amount of the effort over the past 7 years has been dedicated to making certain the kernel can run on as many architectures as possible. Apparently, Sun has decided to pigeon hole Linux as an intel only kernel.

  136. Re:Linux Running on IBM Solaris Running on Sun by shokk · · Score: 1

    So, what you're saying is that less distributions are better? Suppose Sun actually makes a distribution that rocks and blows Red Hat and friends out of the water? We shouldn't have it because you would rather see companies working with Red Hat and Suse? Sun actually has a good patching and patching tracking system in Solaris and Linux would only benefit from it. Solaris and Linux already share a common open source application base, so they are being blurred more and more.

    Frankly, I'll take Sun hardware and Solaris over Linux on any Intel box any day. Sure, you can throw Linux onto these super boxes that have been custom built to supercomputer stats, but most sysadmins are not the University of So-and-so to build your own servers and the Linux business app base is still too small for me.

    Less=Better sounds like something Bill Gates might say.

    --
    "Beware of he who would deny you access to information, for in his heart, he dreams himself your master."
  137. Excedrin by Nemesus · · Score: 1

    This is no different than those pain reliever commercials / styles. Both play with this 'Active Advertisement' games and just prove that they are no better. All they do is say just how bad the competition is but never truely help their case at all. You know ... there is something to be said about those that promote themselves; acknowledge their strengths & weaknesses, and let everyone else know... now they are truely favorable in the public's eye.

  138. Linux is dying by Anonymous Coward · · Score: 0

    Every vendor that has gotten into Linux has
    gotten out or regretted it---Corel,
    VA WhatAmICalledToday, SGI, Dell, Red Hat My
    Stock is Worth Less Than Toilet Paper, etc.

    Sun is right that *linux on mainframes is the
    dying gasp of a dying operating system on a
    dying hardware dinosaur of a platform.

  139. ...where the Sun don't shine... by Anonymous Coward · · Score: 0
    Mr Khan says "Linux on the mainframe just doesn't compute" citing the following reasons:
    • Linux on the mainframe is actually hosted by another proprietary operating system.

      So what if Linux isn't hosted natively? OS emulation has existed for a long time elsewhere. And yes, implementation decisions can rarely be made with foresight into the hardware platform and/or operating system that an OS might be coerced to run on. Sun is just pushing [their] native solutions.

    • Linux on the mainframe is complicated.

      Sun doesn't realize that an IT organization that already has the skill set (mainframe + Unix), is well-suited to Linux on the mainframe. And for an organization whose skill set is primarily in mainframe support, Linux offers skill development/upgrading in Unix.

    • Linux on the mainframe can't respond to the workload demands...it cannot dynamically add resources to match demand.

      System performance monitoring and notification, and some well crafted scripts would allow an organization to increase or decrease the number of images running to reflect demand. And if utilization is measured in this way, utilization is kept high, and it scales with workload demand.

    • Linux on the "mainframe" is not an open system.

      WTF?! However, upon further reading, it's really z/VM which Sun argues is an open system. Well, neither is Windows. What does that prove? This does not preclude IBM from making later advancements to z/VM ... transparently to Linux.

    • Applications that run on Linux for Intel need to be recompiled and recertified for each new platform.

      And his point is --- ??? This is not new. Our applications have to be recompiled and recertified for AIX, HP-UX, Linux (on Intel), and (ta-da) Solaris. If I want cross-platform binary interoperability, I'd use something like Java. Is Sun saying they'd refuse to certify a Java VM for Linux on the mainframe?

    • The economics just don't work.

      The point is, an organization with existing mainframe applications either (1) has an existing mainframe (that maybe under-utilized, and could thus run some Linux images), or (2) is upgrading to another mainframe (to preserve their software investments).

      And if we're talking about the cost of support and services... well, arguably, that may be the future of our industry.

    • Why consider...? Why put...? Why consolidate...? [choose] Sun.

      I found it amusing that the article comes under the heading "Reality Check". This isn't an opinion. It's marketing hype for naive media "reporting" and baseless hooplah for the clueless investment community.

  140. Arrogance by Ogerman · · Score: 2

    Sun is beginning to remind me a lot of Apple. Their hardware is way overpriced and they refuse to accept the fact that commercial proprietary *nixes are quickly dying out, being replaced by Linux and *BSD. Instead, they're sinking large amounts of capital into maintaining their own closed operating systems. It's a shame, really, because they both make well-built hardware, use non-Intel cpu's, and have a solid customer support record. They could be making more money and doing the OSS community a great service by helping out if they'd just wake up.

    1. Re:Arrogance by alex_ant · · Score: 1
      Sun is beginning to remind me a lot of Apple. Their hardware is way overpriced and they refuse to accept the fact that commercial proprietary *nixes are quickly dying out, being replaced by Linux and *BSD. Instead, they're sinking large amounts of capital into maintaining their own closed operating systems. It's a shame, really, because they both make well-built hardware, use non-Intel cpu's, and have a solid customer support record. They could be making more money and doing the OSS community a great service by helping out if they'd just wake up.

      OS X is at least half open-source. Apple's hardware is more expensive than PC hardware, but more expensive != overpriced. You get what you pay for, and Apple these days is not having a hard time getting people to pay for their machines.

      How is Apple's "commercial proprietary *nix" "dying out" when it has already catapulted far beyond Linux and all the BSDs combined on the desktop, in about 1/10 of the time it took for Linux to get where it's at now? How is it "dying out" when, that said, it's currently growing faster than any Unix ever has?

      Do you think the only reason Apple, Sun, IBM, SGI, etc. maintain their own Unixes is because they haven't yet "woken up"?

      Alex

    2. Re:Arrogance by Teutates · · Score: 0

      Apple not giving back? Darwin is completely open sourced, qtss is also, openplay, cdsa (data security architecture), and a slew of documentation are all available on opensource.apple.com...

      You don't know what you're talking about. Apple's hardware isn't that over priced anymore, their software kicks the shit out of many of it's competitors. It's based on open source technology, and they give back. What's your problem?

    3. Re:Arrogance by Ogerman · · Score: 2

      How is Apple's "commercial proprietary *nix" "dying out" when it has already catapulted far beyond Linux and all the BSDs combined on the desktop, in about 1/10 of the time it took for Linux to get where it's at now? How is it "dying out" when, that said, it's currently growing faster than any Unix ever has?

      The concept of proprietary *nix is dying out because it doesn't make sense. People don't use MacOS X to get access to *nix on Apple hardware. Well.. some do, but they quickly realize that it just doesn't cut it, compared to Linux or OpenBSD.

      Now to dismantle your clueless BS:
      #1.) MacOS X is NOT more popular than Linux and BSD, even just on the desktop. How so? Because Linux and BSD users aren't counted whereas people who pay for OS X are. There are many times more Linux boxes out there than OS X. Whats more, most people using OS X are clueless Mac users who would have upgraded whether it was *nix based or not.

      #2.) Since OS X was derived from a version of BSD (and a rather old one IIRC), you can't say that Apple produced it in "1/10 of the time as Linux."

    4. Re:Arrogance by alex_ant · · Score: 1
      The concept of proprietary *nix is dying out because it doesn't make sense. People don't use MacOS X to get access to *nix on Apple hardware.

      It doesn't matter why people use OS X - the fact is that they use it, period. How are they somehow not using Unix just because they never see the command line? Is a Linux user who never sees the command line (okay, maybe this is a little implausible, but imagine a way user-friendly distribution) not using Linux?

      Well.. some do, but they quickly realize that it just doesn't cut it, compared to Linux or OpenBSD.

      Doesn't "cut it"? OS X is BSD, and its Unix component, Darwin, is not proprietary. It is just as open source as Linux. Tcsh and bash are there (well, bash is available separately) and you can compile most POSIX-compliant Unix software (INCLUDING X11 apps, as there is a rootless X11 server available for Quartz).

      MacOS X is NOT more popular than Linux and BSD, even just on the desktop. How so? Because Linux and BSD users aren't counted whereas people who pay for OS X are. There are many times more Linux boxes out there than OS X.

      Linux and every other Unix kills OS X on the server - no question about it. But I think on the desktop you'll find a different story. The commonly-held belief is that Macs currently hold about 5% of the desktop market. Assuming half of these Macs run OS X, that's 2.5% of all desktop computers running OS X. I would be very, very surprised if Linux and all the non-OS X BSDs combined held half of this percentage in the overall, non-techie, non-geek market. Certainly inside such areas, the usage of Linux and other free Unixes is much higher, but Linux-using geeks really do make up only a fraction of a percent of desktop computer users. Believe it or not... I don't see how this is so hard to take, with all the soccer moms buying computers from Sears these days.

      Whats more, most people using OS X are clueless Mac users who would have upgraded whether it was *nix based or not.

      Irrelevant.

      Since OS X was derived from a version of BSD (and a rather old one IIRC), you can't say that Apple produced it in "1/10 of the time as Linux."

      It's been available for 1/10 of the time, though. BSD was quite a bit more mature in the early '90s than Linux was, but come on - Linux has had more than ten years to play catch up. It's not like in 1990, BSD wasn't just a tiny shell of what it would become in OS X. Yes, there was Nextstep, but that was just a faint precursor to what OS X would become. To be fair though, you're right, OS X's foundation was partially already there, while Linux was built from scratch. I'm not sure how significant that is, though.

      Alex

  141. Not designed for a VM, yah right by cullenfluffyjennings · · Score: 1


    Nearly every linux runs in a VM of sorts. The bios initialized your super weird ram bus meemory and other exotic hardware to look more or less like a real fast 486. Claming linux was not designed to run in a VM shows a prety fundemental lack of apreciation of modern computer archatectures.

  142. That's half their R&D by kpharmer · · Score: 1

    In my experience it's about 1/2 of their R&D.

    I will never forget how they wanted my project to use E10Ks instead of Sequents a few years back. The Sequents beat the bajesus out of the E10Ks for scalability, so Sun kept telling folks that they had some breakthrough documents that really dispelled the NUMA architecture. When forced to reveal them they were 4 years old and described the alpha imlementation!

    BTW, I've never worked with a hardware vendor who stooped that low before or since. Of course, database vendors are another story thanks to Oracle and Sybase!

  143. I side with Sun on this one. by duffbeer703 · · Score: 3, Interesting

    Let's face it -- few organizations have people with mainframe talent, and those who have them don't have enough of them.

    So you are going to have to "engage" IBM Global Services to run the thing -- probaly a project manager @ $275/hour and a one or two consultants @ $200/hour.

    Add to this the INSANELY expensive hardware and software maintenance charges every year and you are talking about a serious amount of cash for little benefit

    When you consider the alternatives, it makes even less sense. You can buy 100 Sun E220's or 2-processor intel 4U servers for the cost of one mainframe that lets you emulate 20 Linux boxes.

    Mainframes have been on the wane for the last 20 years for good reason -- they are too friggin expensive!

    --
    Conformity is the jailer of freedom and enemy of growth. -JFK
    1. Re:I side with Sun on this one. by chris_7d0h · · Score: 1

      Have you ever heard of education?
      We're lucky that all our Java developers were born with Java skills, otherwise we would have had to use SUN consultants for $xxx / hour and sign a service agreement with SUN to maintain the application they developed...

      Come to think of it, my first words were: System.out.println("ga,ga,ga, Mommy!");

      --
      In a society that believes in nothing, fear becomes the only agenda ~ Bill Durodié
  144. You want to know what I think? by killmenow · · Score: 2

    What do the folks on Slashdot think?
    I think that when I read a report from someone without a vested interest one way or another, there may be a believable word in it. Until then, I choose to do the typical /. thing: respond without actually reading the article.
    1. Re:You want to know what I think? by kberg108 · · Score: 0

      amen brother

      --
      I like things that are sweet and not things that are lame. --
  145. Newsflash! by Anonymous Coward · · Score: 0

    Sun releases story stating their stuff is good and the competition are poo poo heads!

  146. Yes and no... by iPaul · · Score: 1


    All the stuff about Linux being and intel operating system and performance is just FUD. I think the fact that Linux runs on anything from a palm pilot to a z800 is proof of porability. If raw performance were all that mattered, it's really tough to beat Linux on 1-2 processor intel. In fact go to spec.org and compare for yourself. But raw performance isn't the only thing. Neither is bang for your buck. If I bought bang for the buck I wouldn't buy SUN or RS6000.



    One reason to have multiple boxes is separate environments. This is exactly where the z800 solution is great. I give a set of applications and environments (DEV, QA, INTEGRATION, PROD) their own virtual linux boxes. I don't have to maintain a rack of machines, I just have to maintain 1 machine. In fact I probably don't even maintain that because somebody like PRC does it for me.



    The real problem for SUN is that their 1-4 processor line of machines is in danger of being swallowed up by Linux on Intel. Or in this case Linux on z800 as companies swap out a rack of (mauve?) boxes for 1 black box from IBM. With intel hardware I can get it shipped overnight and with mainframes capacity is a phone call away. Anyone else had the pleasure of waiting 4 weeks for a SCSI disk from SUN?



    This is also why SUN wants to make people think that Linux is okay for "edge environments" where people really weren't buying SUN hardware in the first place. Sun wants to keep it out of 'serious' computing because they realize they 1) don't make intel chips and that's going to be the lion's share of the Linux server market and 2) why pay premium SUN prices if it's just another Linux box?

    --
    Leave the gun, take the cannoli -- Clemenza, The Godfather
  147. Frankly by Rogain · · Score: 0

    Since nobody on slashdot has a mainframe, and MABYE a couple of them ever worked on one, I'd say what slashdoters thinks about linux and VM combo would be of fucking little value. However, that does not mean that Sun is something other than a lying, FUDing bunch of self-serving microsoft-stype monopoly wannabes! (with a fat kernel and sludgy OS, a circa 1992 desktop, who's only saving grace is that they produce quality hardware, grossly expensive but you get what you pay for)

    --
    The current Slashdot moderation system is made by gay communists!
  148. Same critisms apply to Sun vs Windows or Linux by t482 · · Score: 2, Interesting

    1. Hard to find techs
    Yes it harder to find technical people who know sun over windows or linux.

    2.Web serving on sun box is a waste of money
    Why buy one big box when many cheaper ones will do...? Save the big box for the Database server. Tux is much faster than anything from Sun.

    3. Not many applications available
    Most of the good sun stuff is the ported open source stuff that you have to go to ibiblio to download. Applications were mostly written on intel (SAP, Oracle) on linux and then ported to sun. Free stuff is almost all available for linux first.

    4. Incompatibility across versions
    The linux distros now have more in common than the big linux distros. They are more posix compliant as well. For example Sun threads and sun libraries are not portable to AIX or any other OS.

    5. Propretary hardware
    Yup - have a look at Suns network cards... nice but they ain't standard pci cards. 5,000 each for the nice ones. I have found the big vendors scr*w you at every turn. Memory upgrade - 50x the x86 price...

    6. OK for database servers I still see the need. That is until postgresql or some other db vendor supplies fast distributed rdbms.

    Actually I don't have anything against sun - they are probably one of the more "open" unix vendors. But they play the game just like everyone else.

    my 2 cents worth................

    1. Re:Same critisms apply to Sun vs Windows or Linux by r7 · · Score: 0

      >Yes it harder to find technical people who know sun over windows or linux.

      Not in my experience. It is harder to find people with experience managing production Linux servers. The vast majority of Linux resumes we receive are from people with no formal Unix training and relatively little production experience.

      >Why buy one big box when many cheaper ones will do

      Because we don't want to pay someone to upgrade the kernel every few months, or to spend half of every day fixing bugs or debugging hardware problems. The nice thing about Sun that Linux lacks is plug-and-play. Linux on x86 is relatively cheap initially but far more expensive on-going.

      >Most of the good sun stuff is the ported open source stuff that you have to go to ibiblio to download

      And most Linux stuff isn't? You only need to run Veritas or SDS on a terrabyte disk farm to see the advantage to Solaris software.

      >Incompatibility across versions

      Gasp. We're using Sun apps compiled over 10 years ago. Linux, in contrast, seems to break half of everything every upgrade.

      >Propretary hardware

      Thank God, and open-spec too. The best of both worlds. No interrupt conflicts, no incompatible PC BIOS settings, no video card requirement, serial consoles, ...

      r7

  149. Solaris is free by gkhopper · · Score: 1

    Actuall, Solaris is free, as long as you run it on a box with 4 or fewer CPU's, and since Solaris 8 will run on Intel, it's nearly as good as Linux. And if you've ever run Solaris 8 (or 9), you'd almost think that it was a Linux machine (unless you run uname). Solaris 9 will even run Linux binaries (even on Sparc). Oh, and perhaps the reason that Solaris 9 isn't made for x86, is because it would be Linux. So instead of spending $$ to develop something that they can't sell anyway, they suggest that you use Linux on your x86 boxes, (and then buy an e10k to stick in the middle of your Linux cluster). Sun is a big supporter of Linux, they just can't get it to work on their big boxes. IBM has clearly done their homework here.

  150. hehe by Anonymous Coward · · Score: 0

    what this is really about is solaris. Who wants solaris when you can have linux? I wouldnt take competitor's opinions to heart. But anyways, I actually do like sun, and their products. They make good shit.... Still, who gives a fSck what sun thinks?

  151. Goodbye Sun, been bad to know ya,... by RonVNX · · Score: 1

    Nice to see Sun squirm. I won't miss them when they're gone.

  152. Re:I actually work at bioware.. by Anonymous Coward · · Score: 0

    Ok this is very very offtopic, but since your email is @bioware.com,

    How's Neverwinter Nights coming?
    Still March 1 realease date?
    Linux version gonna come packaged with the win32 version?

  153. Re:Well I think Linux SUCKS MY ASS! by gotak · · Score: 1

    News for Nerds werid sex fetish.

  154. Pot Kettle? Blackness? by Anonymous Coward · · Score: 0

    "Recently, IBM announced a new 'Linux-only' mainframe, the z800, which IBM is promoting as a way to consolidate multiple Linux and Unix[r] servers(1). Running Linux on a mainframe doesn't change the fact that you must still maintain an expensive, proprietary system, defeating the whole purpose of introducing open standards like Linux. Although it's technically possible to configure such a system, the question remains, "How well-suited is the system to the task?"

    --snip--

    When I can run something other than Slowlaris on one of Sun's E(pick a number) behemoths, I would be more inclined to take this missive seriously.

  155. Computers literally falling over by billstewart · · Score: 2, Funny

    There's a story from the '89 quake that somebody placed a service call to Tandem after the quake. (Tandem, later bought by Compaq, made highly reliable fault-tolerant machines with N processors, disks, backplane busses, etc. backing each other up.) The machine had fallen over on its face. It was still running just fine, but they wanted Tandem to come put it rightside up just in case anything went wrong in the process... (Also, they were big heavy machines....)

    --

    Bill Stewart
    New Fast-Compression-only CPR http://preview.tinyurl.com/dy575ks
    1. Re:Computers literally falling over by mOdQuArK! · · Score: 1

      I got to do some systems-level programming on Tandem's so-called "fault-tolerant" machines (with 4 pairs of CPUs) for a couple years; it was the standing joke in our department that the machines HAD to be fault-tolerant, 'cause they had to tolerate a lot of faults. One of the CPUs would hang every week or so, and when it did we'd all have to cross our fingers & hope that it would reboot & restart all of the processes for that CPU before the extra load on the other CPUs caused THEM to go down like dominos.

  156. and in another story... by abdulla · · Score: 1

    "Sun Broadens Support For Linux" how do they manage to be both for and against, you'd think they'd make up there minds!

  157. The Way I Feel About The Linux Kernel by Anonymous Coward · · Score: 0

    The fact of the matter is that the VM in the Linux Kernel (2.4 specifically) is "broken". How many different variations of a VM are we going to code? No one can decide which path to take, and it only harms Linux AND The Open Source Movement as a whole. You can't deny the fact that the Linux Kernel is becoming outclassed because the develpers aren't working together to find the best solution to the problems that we face (VM, etc.).

    It sucks, I know, but we need to move past all the petty bullshit and work as a Team, utilitzing all of our resources to create a better Linux Kernel. What that means to me is that every Open Source developer who is devoted to the Linux platform needs to give their input and ideas to the Kernel developers, if not their own code. Otherwise the Linux kernel will continue to fall behind. Guidence is key!

    Don't ignore these facts! Benchmark the latest Linux Kernel against other operating systems and you will see that Linux is no longer ahead of the game in a lot of ways.

    The Open Source movement relies on innovation from anyone and everyone, but at the same time, if developers can't work together then Open Source hinders our cause.

  158. There are other options by Anonymous Coward · · Score: 0

    www.news.com published an article about Virtuozzo technology of www.SW-soft.com - http://news.com.com/2100-1001-843544.html - seems to be similar to IBMs Linux on the mainframe, yet not so expensive and without many of LinuxOnMainframe pitfalls - f.e. it is on Intel hardware, it is managed in the same way, all of the compiled for Intel software just runs, effective sharing of resources and so on. They claim to run over 2000 Linux instances on the conventional 8CPU box from Dell

  159. april fools day already ? by Anonymous Coward · · Score: 0

    [yes, I should look up my account one day]

    Do they know what they are talking about ?

    Note: loads of sarcasm, so think lots of ;)

    >Running Linux on a mainframe doesn't change the fact that you must still maintain an expensive, proprietary system
    Yes, I need some staff (around 1 person per 50-100 z800s) that knows how to handle VM. But on top of that I could run hundreds of linux images (per box -> 50.000-100.000) that could be adminstrated by dummys (hey, every student today has his own linux box at home and most of them know what they are doing.)

    >Compared to z/OS, z/VM is a niche operating system

    hmm .... in the last 10years I haven't seen z/OS (aka OS/390 aka MVS) running without VM underneath it; but wait there are usually multiple z/OS under each VM, so yes, there are more z/OSs running than VMs.

    >with virtual machine (VM) support for new hardware features added late or often not at all(3)

    hmm ... for z/OS VM looks like the raw hardware, so how is that possible ?
    IMNSHO hardware features are implemented in VM long time before z/OS folks even know about them.

    >This is Linux. It's designed for Intel.

    Oh ? That's knews to me.
    If linux was designed for (instead of on) intel, I'd never even think about using it.
    (Hey. Even my carradio runs linux and that's not Intel, but wait it's strongarm. OK.)

    >It's not tuned for the mainframe hardware in which it's running.

    It isn't today, but the guys in the IBM boeblingen lab are moving at a very fast pace. Just look at their kernel updates. And I'm sure there is lots more to come.

    >Despite IBM's claims of easy management(5), customers still need a special machine room and specially trained staff for both z/VM and Linux.

    A server room. Ohmigod. I always have my rack of sun servers under my desk. Makes it warm even through our siberian winters.

    >Horizontally scaled Linux farms are designed to handle unpredictable demand with above average peak loads. As demand rises, a load balancer distributes the traffic evenly across servers, which increases utilization. Because design capacity needs to handle peak demand, server farms often have a low utilization.>Given the relatively low cost of hardware, some organizations find this trade-off acceptable to ensure appropriate service levels.

    Could someone (a native (sun-)english speaker) let the air out of this and tell me what's the point ?!

    >Although z/VM can start and stop Linux images, it cannot dynamically add resources to match demand.

    If you run out of CPU-power, call IBM and let them enable an additional CPU ? Usually takes a few minutes; And if you have a good contract they'll do it on the fly, without you calling them...

    >and there is little incremental RAS benefit.

    Reliability. I never saw a full break down of a zServer.
    Availabilty. After you saw an IBM tech turn up at the same time as the UPS truck, telling you: "Ahh there are the replacement parts for your broken hardware..." - "We had broken hardware ?!". You know SUN is bsing.
    Security. They are the last ones to talk about security.

    >For example, z/VM cannot take advantage of Parallel Sysplex clustering, and VM hypervisor is an added single point of failure(8).
    >(8) According to Sun technical experts

    That's a good reference. I guess Sun (/solaris ?) experts know what they are talking about.

    >Applications that run on Linux for Intel need to be recompiled and recertified for each new platform; thus the application portfolio to run Linux on a mainframe is small(9).

    Yes. A recompile for a different hardware-platform takes so much longer than a port to a totally different OS (Solaris). Maybe if you don't know how to write code/portable code. But then, who wants to run his server on some VBS-Kiddies code anyway ?

    >Often the difference in Intel versus mainframe applications makes porting difficult(10).
    >(10) Intel uses something known as little endian; a mainframe uses something different. This is significant for certain applications and makes the port difficult.

    Little versus big endian makes porting difficult ?
    Did they ever _do_ a port ?
    In our code we don't even care if the data is represented in 0s and 1s or apples and oranges.

    >Additionally, different applications are ported to different distributions of Linux (for example, Red Hat, SuSE, and Turbolinux). Getting applications to run on the mainframe might require supporting multiple distributions of the 'same' Linux operating system.

    Everybody who ever wrote a linux (intel) application will know that this is hot air.
    Sometimes dynamic linking makes things a bit harder, if you require bleeding edge libs, but that's your fault anyway.

    >Sun's rack-optimized 1U form factor servers start at list prices under $1,000.

    I've been searching for a 1U box under $1,000 for quite some time now. This box they are talking about couldn't even handle my server at home (2 persons, firewall and answering machin, plus some fun stuff.) Talking about apples and oranges again ?

    >Thus, when considering consolidation projects, why consider putting workloads onto a completely different and more expensive architecture?

    No, use solaris intel. At least it's the same hardware. erm....

    >Why consolidate on a system with limited application support and one that demands a rare combination of skills?

    And I thought they wanted to sell those sun boxes.

    >Sun provides:
    > * Binary-compatible architecture

    Binary-compatible with what ?

    > * Over 12,000 available applications
    hmmm ... and 11.500 of them are hello world variations.

    > * Servers that scale vertically to 106 processors and are sized and priced to fit any task

    huh. 106. That is even more than an intel wolfpack, isn't it ?

    > * Support for open standards and Java[tm] technology

    Oh yes. And the linux/390 vm is just slightly faster, but that's not open standards, since it runs on closed hardware....

    > * Linux compatibility features of Solaris

    Yes, linux compatibility is much better than native linux.

    > * World-class dynamic provisioning capabilities and management tools

    Ever saw them in action ?!

    > * Dedicated Sun Professional Services consolidation practices

    Go and ask their customers about this one ....

    Note2:
    This is just my personal view. Not that of my company.
    And not that of my Manager. (BTW: I don't work for Shahin Khan).

    anti

  160. well, why bother about the mainframes... by seppuku · · Score: 1

    ...when Dell's plain Intel SMP box could run 2000 copies of Linux? Check this - http://news.com.com/2100-1001-843544.html.
    It looks much more clever and efficient than VMware virtual machines, even if it "limits" you to one OS - Linux.

  161. The link is a good article! by Bo+Vandenberg · · Score: 1

    I'm not saying I agree with it, frankly I don't have enough mainframe exp. to be informed, but as a supported argument it's compelling.

    Its hard to refute that the solution is a young one. I would love to see a technical support of the VMos that orchestrates all the virtual machines.

    $400k and expensive repairs for a mainframe verses around $20k for 20 linked 1u rack mounted computers is pretty compelling too.

    Size of installation doesn't seem much of an issue with stuffed racks of 1u boxes. I'd love to see a measurement\visualization.

    The author of the report, footnoted all his points and gave much better technical support than 90% of the posts here.

    Honestly, the reply posts overall, with some exceptions, don't speak well of our little forum: too many people posting flames, ignorant posts from people who probably have no mainframe experience, and a definite "my OS is bigger than your OS" mentality. -- This makes me sad. We are better than this, and more useful when we want to be.

    Bo

  162. You do not know what you are talking about! by beezly · · Score: 1
    no reason why you cannot run multiple, different OS' on either Serengeti (3800, 4800, 4810, 6800 boxes) or Starcat (SF15K).

    I do this every day of the week. All the multi domain Sun boxes are quite happy running whatever version of Solaris you want and there is no technical reason why you couldn't mix Linux/Solaris/(some other Sparc64 OS) on different domains at the same time.

  163. Marketing Schmarketing by Anonymous Coward · · Score: 0

    Since when did the "Marketing" lot, ever really know anything anyway?, its all fluffy footballs and lovely girls isn't it?

    Sun does have a point, with regards to maintenance and support for an expensive platform, but don't they shoot themselves in the foot?...they aren't selling cheap boxes anyway.

    If you have an old mainframe (wink), and want utilise it better, and VM will run on it, that sounds more cost effective. Although if you company is paying for maintenance/support etc, then I doubt they would mind upgrading.

    Could you imagine someone hacking into an IBM Linux/VM box?, the power, the control!!

  164. Don't any of you GET what Linux is and can be? by cbreaker · · Score: 1

    This article is interesting, if not completely biased due to the origin of.

    I don't know either way wether mainframes and Linux mix, at least in this manner, but it's not a bad thing. Nothing that pushes linux further is a bad thing.

    I read through so many posts on Slashdot, and I see a lot of people that just don't seem to "get" it.

    Linux is great for the computer world in general. It's free, open, and powerful. It's gaining new ground and userbase by the day.

    It scares the crap out of the competition. How do you compete with something that is free? That is open source to everyone? And what happens because of this? You have to make your offerings better, cheaper, etc. to keep up. That's good!

    Most of you don't need to run it. You can run Windows XP and be happy. That's fine! Run whatever you want to run, it's your machine.

    So many people say "Linux is for this." "Linux is for that." Well, it's for whatever you want to do with it. You want to run a small web server on Linux because of the robust and flexible nature of UNIX? Do it! You want to run a powerful 8-way SMP machine to do heavy processing? It's being done now. You can do that too. Want to run Linux as a workstation? You can do that too.

    If you want to run software that is not designed for Linux, well, too bad. Just like anything, use the system that suits your needs. If that is a Mac, use a Mac.

    I agree that Linux isn't as user-friendly as a Windows or MacOS based machine when it comes to the workstation side of things. It's not there yet. Boy, has it improved on that front though! Someday a first time computer user will have just as many problems, hopefully less, than someone who is trying to install Windows. :)

    Anything that pushes Linux further is a good thing. I hope someone does feel that this product from IBM is useful to them, and promote more innovations in the future from IBM.

    So why do people use Linux? Because they choose to. It suits their needs, or they like the control over their systems that UNIX provides. Someday, I hope that the average computer user will also use it because the software they want to use runs on linux, and it's easy to use.

    Why is it good for the industry? It's free, powerful, and only going to get better. Linux is obviously getting huge. Isn't it wonderful to see products from so many different competitors all sharing a common system? Who knows, it could even lead to true compatibility some day.

    Well anyways, think twice before you judge anything. Whether you run it/like it or not, you can't ignore what could be the biggest evolution in the computer industry. And the evolution isn't even Linux itself, it's the idea of free, open, and very good software that works.

    --
    - It's not the Macs I hate. It's Digg users. -
  165. Hack it by Anonymous Coward · · Score: 0

    Scary thought, maybe, what if you hack VM, and take control of the mainframe. The power!, through Linux.

  166. Endianness /is/ a problem. by himi · · Score: 2

    That said, anything that will compile and run properly on ppc (big-endian) as well as x86 (little-endian) should be endian-clean, and hence should run fine on a mainframe.

    himi

    --

    My very own DeCSS mirror.
  167. x86 Linux emulated? by Thaidog · · Score: 0

    Please... Do you really think IBM would emulate x86 Linux on a mainframe? God, I thought I was a cracker. Anyway, it's called OS/390 Linux. It's made for... OS/390's and you can pick it up at SuSE or RedHat's website. The great thing about these systems is that you can run many operating systems on one box. I'm not too sure, but I'm guessing z/VM is what allows this. And no, it's not an emulator. Why would you emulate x86 Linux anyway? Why not LinuxPPC for that matter? x86 Linux is like DOS made to run on dirt cheap hardware. Would you like to see what Linux can REALLY do? Check it out on an os/390...

    --

    ||| I still can't believe Parkay's not butter.

  168. Moderators! by Eagle7 · · Score: 2

    Mod the Parent *DOWN*.

    I've developed Linux on the mainframe, and none of this is true. It IPLs very quickly, and is just as easy to use as any other Linux. That post ought to be (-1, Troll), not (5, Informative).

    Here for proof that I know what I am talking about (down toward the bottom).

    --
    _sig_ is away
    1. Re:Moderators! by Dixie_Flatline · · Score: 2

      Moderators, ignore this man.

      He may have developed on linux on an IBM mainframe, but it's clear that he's never tried to maintain such a system, or worked in an Administrative capacity. I have no possible reason to make this up.

      In any case, one good experience doesn't constitute proof that linux ALWAYS works on a Mainframe. Likewise, my experience doesn't constitute proof that it's always going to be a pain in the ass.

      It IS just as easy to use as any other linux, it just doesn't run as well.

      Lastly, those screenshots don't PROVE anything, per se, though I don't see why he'd fake them.

  169. Why I'd hire the Linux Advocate by leonbrooks · · Score: 2
    You go apply for a job and put "linux advocate" on your resume and I will put advocate of "quality computing" on mine.

    I'd hire the Linux Advocate because at least he doesn't have his head up his ass so far that he can't see out (ie no delusions of grandeur).

    Many, even most quality systems suck like a Kirby: they fail to deliver on their promises, introduce significant overheads, and cost heaps. It's a good way to dig yourself into a hole with confidence.

    Quality Consultant is usually resume code for ``I'm useless in my industry, so the only job I could get was as a combined critic, form-herd and bullshit merchant.''

    The Linux Advocate will actually have some focus, and will be more likely know enough of the details of what he works with to be useful.

    I can do everything linux can on Solaris

    Really? Write or do significant modifications to a device driver then, go on! A driver for Conexant WinModems would be a good place to start...

    --
    Got time? Spend some of it coding or testing
    1. Re:Why I'd hire the Linux Advocate by Anonymous Coward · · Score: 0

      /I I'd hire the Linux Advocate because at least he doesn't have his head up his ass so far that he can't see out (ie no delusions of grandeur).
      /i

    2. Re:Why I'd hire the Linux Advocate by Anonymous Coward · · Score: 0

      I would see "Linux Advocate" as "Linux Zealot", if you hire me I will try to put Linux on every computer in your company wether it makes sense to or not, and I will refuse to learn or work on anything that is not Linux. No thanks.
      Critizing an OS because you can't write/modify a device driver for it is pretty lame. One of the reasons linux sucks so bad is because you HAVE to modify the source code just to get the crap to work right, NO THANKS! :-)

    3. Re:Why I'd hire the Linux Advocate by Tet · · Score: 2
      I can do everything linux can on Solaris
      Really? Write or do significant modifications to a device driver then, go on! A driver for Conexant WinModems would be a good place to start...

      Easy. Sun have provided the ability for third parties to write drivers for Solaris for as long as I can remember. See here for everything you need to know. If you're going to slate Solaris, at least base it on facts. I have been using both Solaris and Linux for over 10 years now. I can do almost everything on Solaris that I can do on Linux, and I can do almost everything on Linux that I can do on Solaris. Both are good operating systems, and each have their strengths and weaknesses. Use whichever one is right for you for the task in hand, and be thankful you don't have to use Windows...

      --
      "The invisible and the non-existent look very much alike." -- Delos B. McKown
  170. Sun's real problem by kune · · Score: 2

    Imagine a CIO of a Fortune 500 company, planning a new web project. For the hardware he has two options:

    /1/ Add a dozen or more CPUs and some RAM to the existing mainframe infrastructure, to support a number of Linux images. (His data center probably runs a number of several MVS, OS/390 and maybe z/OS images in parallel.)

    /2/ Buy and install two new Sun Starfire 15K.

    Obviously his choice would be /1/, if the software guys are saying that it will work. Probably they will have all a Linux partition on their PC :-).

    If costs is the strong argument in the situation, Sun will have only a chance with large discounts.

    I think Sun's strategy to build the next mainframe with the 15K backfires here, because for web servers the Intel rack hardware running Linux is the better option and IBM's virtual OS images will use CPU and RAM more efficiently than Sun's domain per 4-processor cartridge by definition. The nightly mainframe batches map well with the usage patterns of online services.

    So Sun's revenue is now attacked by Linux from below and from above. These guys are forced to love Linux. Remember Linus's joke about world domination?

  171. Informed bashing?! by korpiq · · Score: 2

    I took a quick look at the article and was astounded. It's to the point and seems to address real concerns - at least to one like me who has no opportunity for first-hand testing.
    These points really require addressing from IBM's side, I think. Well, waiting for that to appear on /. frontpage soon :)

    There was no article rating system available on Sun's page, so I had to write this here :) Kudos for the writer. Bashing, but informed instead of empty FUD.

    On the other hand, free OS'es certainly don't grow on mainframes; they grow in the wild, so it's natural they're more fitted to the small-to-medium size HW. It doesn't lessen their absolute importance as cost-efficient clusters (Imagine a Beowulf of these! ;) ), compartment servers, home machines, firewalls or embedded solutions. Even more, it feels to inversely prove their fittedness for such tasks, as they're obviously "by their very nature" more geared toward such environments.

    OT: It's been a nightmare with the Kernel of Pain for me even on Intel hardware trying to put up some more serious servers (I wont' go into that in length, but reiserfs/RAID losing content still in 2.4.17 - blah!). I'm sticking with 2.2 stables and taking time to apply frisbee on server side in the future.

    --

    I think, therefore thoughts exist. Ego is just an impression.
  172. Sun distributes Linux servers by konmaskisin · · Score: 2

    and they're great:

    http://www.cobalt.com/

  173. I would have read that article, but... by Fefe · · Score: 2, Insightful

    their web server times out on me. Why?
    Because Sun's firewall is broken and drops TCP connections using Explicit Congestion Notification (ECN).

    How stupid can a single company be? I bet that article goes to great lengths to say how well suited Sun is to provide scalable web servers. And they can't even get their own web server configured properly!

    What a buch of losers. ECN is, by the way, an official internet standard (RFC3168), which happens to be implemented by Linux.

  174. Re:I actually work at bioware.. by Dixie_Flatline · · Score: 1

    The best place for information is still on our webpage, on our forums. I don't actually work on the Neverwinter project myself, so I can't really answer your questions. In any case, I don't think we ever promised a March 1 release date...not in recent history, anyway.

  175. Sun isnt the stirling saint by hesperant · · Score: 1

    While it is true Sun seemes to have pioneered many of the technological advances in Mainfraim computing. They have not gone far enough. You dont stop and say.. There .. i'm Done when you have doubled the abilities and then bash on the competition hoping to create doubt with ill concieved references and incomplete information. That is what almost landed IBM in the doghouse and sent Digital to visit the briny depths. Who cares what they "Did" it's what they are "Doing" that counts. The 64 bit architecture cannot stand up to todays base 32 bit architecture in business applications unless you use a proprietary OS and maintinance is maintained through a costly contract. IMHO Sun put together the idea of sticking an off the shelf Distrobution into a mainfraim service and found it wanting. The best part is the fact that they did this.. it demonstrates the versatility of Linux and the concept of open source, by proving that it can be done without proprietary equipment. Lets also add. that In a hands down test The linux based Beowulf Clustering system witch can be labeled "mainfraim" as the concept of mainfraim verses super computer is a matter of orgonizing the equipment has been able to best most of SUN products in the big race for computation and these systems use a combination of older 32 bit based machines. IMHO Sun stopped.. they have run into a wall and are not looking for a way over it.. they are looking to add the same wall to someone elses landscape and dont realize how Ironic that is.

  176. Dynamic resource allocation for Linux VMs ??? by zaqattack911 · · Score: 1
    An interesting quote from the article was:

    Although z/VM can start and stop Linux images, it cannot dynamically add resources to match demand. As a result, a mainframe would need to size for peak demand just as the Linux farm would; high utilization is a myth.


    Is that true? And, if it is... it's a good point I would think. If multiple linux server images are to be the new fashion of servers, what would be involved in having resources such as ram and diskspace allocated dynamically based on demand of one of the linux images running?

    On a final note, I'm starting to really think this whole multiple linux images running on one machine trend to be just a "fad". In most cases (granted not all), isn't it just as effective and easier to have all the services running under one linux server? maybe some extra magical software to control resources etc?
    --Me

  177. In other news by Anonymous Coward · · Score: 0

    Microsoft claims Internet Explorer is superior to Netscape Navigator.

  178. The anit-FUD is FUD by AKAImBatman · · Score: 2

    You also get capacity upgrade on demand with a phone call with the IBM hardware. They dont even have to send out a CE to do anything. Let's see SUN do that.

    Sun does that too with their big machines. The scheme is called COD (Capacity on Demand) and works the same as IBMs plan. Also, to reply to the other reply to your reply (confused yet?), Sun machines are rather unique due to the fact that they can run both endians, big and small. I've never actually heard of anyone DOING something with that feature, but it does exist none the less.

    1. Re:The anit-FUD is FUD by Jay · · Score: 1

      Both endians?

      Schweet! I dunno if I believe it, but it's still a really cool idea.

      --
      You think emacs is evil?! You've never used VM's XEDIT have you?!! That's evil, baby!
  179. Grasping at straws... by Anonymous Coward · · Score: 0

    To someone with a cursory exposure to mainframes running z/OS and z/VM, it's blindingly obvious that Sun's web page is just an attempt to FUD the mainframe. One of its base assumptions is that z/VM is supposed to be an OS for serving applications, ignoring the fact that on zSeries, z/VM is practically relegated to the narrowly-defined role of "hypervisor."

    This kind of FUD, with all its half-truths and ignorant assumptinos, is almost worthy of Microsoft.

  180. Pote boudas pote koudas by Commienst · · Score: 1

    Song: Pote boudas pote koudas
    Artist: Nikos Papazoglou


    Pote Voudas pote Koudas,pote Ihsous ki Ioudas

    Exw katalavei hdh ths zwhs mou to paixnidi

    Exw katalavei hdh thw zwhs mou to paixnidi

    Pote Voudas pote Koudas,pote Ihsous ki Ioudas

    Olo idia kai ta idia tou myalou sou rokanidia

    Vre den einai edw to Souli,edw einai tou Rasoulh

    Vre den einai edw to Souli,edw einai tou Rasoulh

    Olo idia kai ta idia tou myalou sou rokanidia

    Sto 'pa mia kai sto 'pa dyo,sto 'pa xilies deka-dyo

    Allo o anoixtomaths ki allo o avgoulomaths

    Allo o anoixtomaths ki allo o avgoulomaths

    Sto 'pa mia kai sto 'pa dyo,sto 'pa xilies deka-dyo

    Pote Voudas pote Koudas,pote Ihsous ki Ioudas

    Exw katalavei hdh ths zwhs mou to paixnidi

    Exw katalavei hdh thw zwhs mou to paixnidi

    Pote Voudas pote Koudas,pote Ihsous ki Ioudas

    --

    I am into the copy and paste.
  181. Re:Sun has lost...(because of IBM) by Anonymous Coward · · Score: 0

    while ebay has, instead of running away at one bug on the sparcII hardware (which is fixed in the new mirrored cache modules, and doesn't exist in the sparcIII CPUs), stuck with Sun and is now highly pleased with their systems.

    It's also interesting to note that the SRAM chips that were buggy on the ultrasparcII CPUs were made by IBM, and that IBM employees know about this little bug when they sold them to Sun... funny they didn't mention it, no?

    Sun is taking full blame for not checking it out better, but I think it's a pretty interesting situation.

  182. Or... you could have it never work right by leonbrooks · · Score: 2
    Apologies to the moderators for replying to a Score:0 AC comment, but one point does bear answering.
    One of the reasons linux sucks so bad is because you HAVE to modify the source code just to get the crap to work right

    Firstly, I have to modify Linux source code very seldom (maybe one system in thirty), and when I do it's for very obscure reasons, and it's good (yea, verily, even great!) to have the choice in that circumstance - especially since I specialise in odd systems (the kind of stuff that the MCSE brigade don't even dream of doing (and when they're stupid enough to try anyway the resulting house-of-cards heap of VB makes Heath Robinson look positively conservative)).

    If I installed ``cookie-cutter'' Office+Web+Email desktops all day, and servers to suit, there's no reason I'd ever bother working with source. Or for that matter, for dirtying my hands with media, since net boot ROMs have become common.

    The alternatives are:

    • It never works properly
    • You have to hope that the issue becomes important enough to the vendor for them to make a fix (and if they're Microsoft, also pray hard that the fix doesn't make things worse).
    • Your company is important enough to the vendor to permit bullying them into fixing it
    --
    Got time? Spend some of it coding or testing
  183. Good response, thanks by leonbrooks · · Score: 2
    Sun have provided the ability for third parties to write drivers for Solaris for as long as I can remember.

    Some drivers, yes. I shouldn't have been quite so specific.

    I can do almost everything on Solaris that I can do on Linux, and I can do almost everything on Linux that I can do on Solaris. Both are good operating systems, and each have their strengths and weaknesses. Use whichever one is right for you for the task in hand, and be thankful you don't have to use Windows...

    Amen, brother, amen... (-:
    --
    Got time? Spend some of it coding or testing
  184. PS, Advocate != Zealot by leonbrooks · · Score: 2
    I will try to put Linux on every computer in your company wether it makes sense to or not

    Whereas a quality zealot will bury you in mostly-useless paperwork instead. Great choice.

    But Advocate != Zealot. I'd avoid hiring a zealot of any kind, if I could, they're more trouble than they're worth. This is not to say that strong personal preferences are bad (they generally aren't), but that monomania is.

    --
    Got time? Spend some of it coding or testing
  185. Hrm.. by Dragnet · · Score: 0

    Looks to me like there is an ongoing war between the SUN article editors...

    Reality Check: Linux on the Mainframe--Not a Good Idea
    This article discusses the recent IBM Linux-only mainframe announcement, and the complexities this z800 introduces into your IT environment.

    Feature Story: Sun Broadens Support for Linux
    Aggressive new program expands the role of Linux on entry-level servers.

  186. Sun Bashes Linux on (IBM) Mainframes by Anonymous Coward · · Score: 0
    What do the folks on Slashdot think?
    Linux on the Mainframe--Not a Good Idea
    By Shahin Khan, Chief Competitive Officer, Sun Microsystems, Inc.
    February 20, 2002
    Of course it's not a good idea - for SUN. But it may be a good idea for IBM, IBM's customers or Linux.
    Linux on the mainframe is actually hosted by another proprietary operating system, z/VM. The optimized operating system for IBM mainframes is z/OS, formerly called MVS(2).
    That's not what the VM bigots say. And z/VM does include facilities that z/OS lacks.
    Compared to z/OS, z/VM is a niche operating system with virtual machine (VM) support for new hardware features added late or often not at all(3).
    Sometimes MVS support for new hardware features was added late or not at all, e.g., FBA.
    And Linux isn't designed to run in a virtual machine; implementation decisions that make sense on PC hardware don't fit well in a virtual machine(4).
    Utter balderdash. MVS wasn't designed to run in a virtual machine either, and it runs perfectly well in an LPAR or under z/VM. ).This is Linux. It's designed for Intel. Well, I guess that means that we should buy Intel rather than SUN if we want to run Linux ;-) Except, of course, that the statement is false. It's not tuned for the mainframe hardware in which it's running. It's not tuned for any specific hardware. Tuning is something that you do after you have installed the software at a specific site for a specific purpose.
    Linux on the mainframe can't respond to the workload demands of Web serving with high utilization--something IBM touted at the time of its z800 announcement.
    This would be relevant if it were true - which it isn't.
    Contrary to what many believe, consolidating a Linux farm into multiple images on a mainframe would not change the demand pattern. Although z/VM can start and stop Linux images, it cannot dynamically add resources to match demand.
    The first part is false and the second part is misleading z/OS and z/OS.e can shift resources among Linux virtual machines.
    It's neither fish nor fowl. Linux on the "mainframe" is not an open system,
    z/VM and the IBM device drivers may not be open, but the Linux Kernel and the GNU utilities certainly are. And the whole package is more open than Solaris.
    and there is little incremental RAS benefit. Although IBM claims "zSeries servers inherit the legendary IBM S/390 strengths in the areas of fault avoidance and tolerance, recovery from failures, and concurrent maintenance and repair for "always-on" availability"(7). We don't believe this to be true for zSeries servers running Linux.
    ITYM "We wish it weren't true.
    The "legendary" IBM S/390 strengths IBM references are the result of decades of development work on IBM's flagship mainframe operating system, known today as z/OS.
    Close but no cigar. See below.
    The fault recovery features of z/OS are not found in Linux. z/VM does have some fault recovery features, but it is not nearly as resilient as z/OS.
    They include alternate CPU recovery (ACR), recovering from CPU, channel and memory failures and hot swapping of CPUs and DASD. Solaris has many more single points of failure.
    For example, z/VM cannot take advantage of Parallel Sysplex clustering, and VM hypervisor is an added single point of failure(8).
  187. Tre varmkorvar ("Three Hot Dogs")? by CRConrad · · Score: 1

    Hmm... And here I thought the T-68 looked so nice.

    Guess I'll have to stay with handsets from that gum-boot factory in the Eastern Half Of The Realm, then.

    --

    Christian R. Conrad
    mail me at iki.fi ; same user ID as here