Slashdot Mirror


The Eternal Mainframe

theodp writes "In his latest essay, Rudolf Winestock argues that the movement to replace the mainframe has re-invented the mainframe, as well as the reason why people wanted to get rid of mainframes in the first place. 'The modern server farm looks like those first computer rooms,' Winestock writes. 'Row after row of metal frames (excuse me—racks) bearing computer modules in a room that's packed with cables and extra ventilation ducts. Just like mainframes. Server farms have multiple redundant CPUs, memory, disks, and network connections. Just like mainframes. The rooms that house these server farms are typically not open even to many people in the same organization, but only to dedicated operations teams. Just like mainframes.' And with terabytes of data sitting in servers begging to be monetized by business and scrutinized by government, Winestock warns that the New Boss is worse than the Old Boss. So, what does this mean for the future of fully functional, general purpose, standalone computers? 'Offline computer use frustrates the march of progress,' says Winestock. 'If offline use becomes uncommon, then the great and the good will ask: "What are [you] hiding? Are you making kiddie porn? Laundering money? Spreading hate? Do you want the terrorists to win?"'"

225 comments

  1. Deep by roman_mir · · Score: 3, Insightful

    Wow, so deep. Computer is the Internet, Internet is the computer.

    Mainframes are specialised equipment, server farms are almost generic computers with redundancies. The real difference is the cost. Today's server farms would cost many factors more if they were built with specialised mainframes, there is no other real difference, they are really there for the same purpose.

    1. Re:Deep by tarpitcod · · Score: 5, Informative

      Right and there are some big differences:

      Mainframe CPU's tend to have far more error detection and correction. They have safeguards against errors in data shuffling and computation inside the CPU itself. Mainframes tend to offer robust job control, by the time you add decent job control of the level that mainframes offer your network of workstations/servers starts getting complicated
      Mainframes tend to offer decent encryption and security.

      Can you do all these things on a pile of VM's? Sure. Is it cheaper - maybe. Is it fun to manage - not particularly.

      For the point about giving everyone access to all your stuff? Let's see the author prove his point by posting all his personal details, address age, credit card numbers, ssn, medical records, tax returns and let's see how that works out for them..

    2. Re:Deep by hairyfeet · · Score: 2

      Well last I checked, and its been awhile so it may have changed, the big difference (which also ramped up the price) is all the extra layers of error checking and failover in a mainframe so you absolutely can be 100% positive you are getting the right answer 100% of the time. which when you consider that they are often used for finances...yeah i can see why that would be of importance.

      But that is why you have a lot more companies using server farms than using mainframes anymore, there is just more work that doesn't have to have five nines levels of precision and commodity hardware means that you can get some crazy levels of number crunching for cheap.

      --
      ACs don't waste your time replying, your posts are never seen by me.
    3. Re:Deep by Trepidity · · Score: 4, Interesting

      I agree these are all differences for a regular pile of VMs in a server room, but if you look at some of the more developed server farms, they do have a lot of the mainframe-like features, at least on the software side. Google, for example, has pretty full-featured job control layered on top of their server farm.

    4. Re:Deep by SuricouRaven · · Score: 2

      From a technical perspective, big difference. From a business perspective, not so much. The business side doesn't care about just how the technology is built. What matters is that mainframes and server farms are a black box in a company-controlled office built with company-controlled hardware where vast amounts of data are stored and processed. Centralisation and specialisation.

    5. Re:Deep by swalve · · Score: 3, Insightful

      That stuff is also in hardware, which is only beginning to happen in the commodity pc world.

      For a certain type of workload, at a certain level of necessary uptime, mainframes start becoming cost effective. Fun things like where IBM will install as many CPUs as you want, but only charge you for their time when you use them. This can be very cost effective for businesses with seasonal volume shifts. At some point, paying IBM $1000 an hour for their support is cheaper than paying 20 creeps with greasy hair to change hard drives, stack servers into a rack and fuck up the rollout of new VMs. It's kind of like trucks versus trains. Each have their place, but neither is very good at emulating the upsides of the other.

    6. Re:Deep by Nerdfest · · Score: 1

      You can get the same precision and fault tolerance by using commodity hardware by running multiple jobs in parallel, but it's rarely required.

    7. Re:Deep by tarpitcod · · Score: 2

      It's rarely required, until it is.

    8. Re:Deep by Emperor+Shaddam+IV · · Score: 2

      Mainframes aren't so "specialized". Maybe you are confusing Mainframes with Supercomputers which tend to be much more specialized and focused towards scientific and research usage.

      I worked on IBM big iron back in the day and a "mainframe" can run Linux Partitions as well as other mainframe OS's. Unix boxes aren't so generic either. A unix box running Linux is different than a Unix box running HP-UX or Solaris and requires some different sys-admin skills. There are other issues with shared library linking being different, different compiler's, different shells, etc.

      MVS is now z/OS and it supports multiple programming languages - its not just your grandfather's COBOL anymore:

      http://en.wikipedia.org/wiki/Z/OS

    9. Re:Deep by Ken+Hall · · Score: 5, Interesting

      I work with mainframes for a living. Specifically, I work with Linux on IBM zSeries mainframe for a bank. The idea is the provide the software depth of Linux with the reliability of the zSeries hardware.

      We get a fair amount of resistance from the Lintel bigots, mostly those who still think of the mainframe in 1980's terms. The current generation of mainframe packs a LOT of horsepower, particularly I/O capacity, in a relatively small box. It connects to the same storage and network as the Lintel servers do, but can one of those do 256 simultaneous DMA transfers? We don't sell the platform as a solution for everything, but we've done the TCO math and we're not that different from an Intel server farm once you factor in the external costs.

      I periodically give a class to the Linux admins on the mainframe in general, Linux on z, and the differences between that and Linux on Intel. If you didn't know where to look, it would take you a while to figure out you're not on Intel anymore. Most of the attendees are surprised at what the current boxes are like.

      This is not your fathers mainframe.

    10. Re:Deep by tverbeek · · Score: 1

      Don't forget the low-cost dumb terminals – I'm sorry: "thin clients" – which are incapable of doing anything at all independently of the centrally-adminstered silicon. The computing environment I work in today is architecturally very similar to the one I started working in back in the mid-1980s.

      --
      http://alternatives.rzero.com/
    11. Re:Deep by DarkOx · · Score: 1

      That and its rarely done because frankly the typical LOB software developer does not know how to implement such things. Lets face it even the cheapest hardware is so good most of the time they don't need to. Its also true they should not have to. By the time someone is writing x = x * y; in Java or even C, other than being sensitive to data-type, will it overflow? is a float that is going to have precision truncated? etc; they ought to be able to depend on that working as expected.
      The right place to deal with the issue really is in hardware or not at all.

      --
      Repeal the 17th Amendment TODAY! Also Please Read http://www.gnu.org/philosophy/right-to-read.html
    12. Re:Deep by ArsonSmith · · Score: 3, Interesting

      I build server farms specifically to suck data out of Mainframes and process it specifically because of the cost difference. It is nearly 100x the cost and still takes 10x longer to crunch, index and search 8PB of data on mainframe as it does in a comparatively free Hadoop cluster. The TCO was laughably different.

      --
      Paying taxes to buy civilization is like paying a hooker to buy love.
    13. Re:Deep by roman_mir · · Score: 1

      I said:

      there is no other real difference, they are really there for the same purpose.

      .

      By 'specialised' I do mean they have more hardware built in to achieve higher levels of data throughput and error correction.

    14. Re:Deep by mikael · · Score: 2

      That's the difference - the traditional mainframe was a one vendor product - racks, disk drives, CPU's, network boards, cables, terminals, everything available from the one supplier at "special" corporate rates, providing that you gave them the exclusive maintenance contract. Want printed system manuals? We'll charge you for that. Want more than eight user accounts? That cost extra too. Need a compiler for OS development work? That'll cost extra. Want the pre-compiled development API's to write applications? That'll cost more too. Want an optimizing compiler for high performance applications? That's cost some more too. Need a cable for your laser printer? We''ll supply that for a fee.

      Compare that to the current server room where everything has generic components from the racks, cabinets to the fans, memory, network boards, cabling. If you consider that you can buy CPU's from any number of suppliers even if they are AMD/Intel, then they too are generic components. Everything removable and replaceable whenever technology advances.

      Must have been 6-7 years ago, but when I was at college, every room had a locked cabinet with three or four router/terminal servers like boxes interwired together. Three years later, the network had been updated, and those boxes had been made redundant as the processing had gone back to the server room.

      --
      Vintage computer adverts: http://www.vintageadbrowser.com/computers-and-software-ads
    15. Re:Deep by Anonymous Coward · · Score: 0

      For the point about giving everyone access to all your stuff? Let's see the author prove his point by posting all his personal details

      {{citation needed}} Where does the author make that point, in your opinion? I think he's arguing against it.

    16. Re:Deep by Nerdfest · · Score: 1

      You haven't tried the IBM kool-aid yet. Those people whose jobs currently rely on mainframe expertise are very happy with them. They do have better error-checking but everything else is at least an order of magnitude out of whack with commodity hardware price/performance, and in many cases, several orders. You can reduce some of the costs on their zSeries by buying specialised processors for DB2, Java, and Linux (~100K a pop) so you don't have to may for MIPS usage but the costs are still astronomical for the performance. If it was cost effective, don't you think Amazon would be running its cloud services on them?

    17. Re:Deep by Nerdfest · · Score: 1

      Ever check the cost of those 'low-cost' IBM terminals?

    18. Re:Deep by ShanghaiBill · · Score: 1

      You can get the same precision and fault tolerance by using commodity hardware by running multiple jobs in parallel, but it's rarely required.

      It also rarely makes sense. If the parallel instances are running the same software, they will likely both make the same error, since 99.9% of reliability issues are in the software not the hardware. If you spend a million dollars on more robust hardware, and a million dollars on extra software testing (unit, integration and (especially) usability), the latter is orders of magnitude more likely to prevent a problem.

    19. Re:Deep by JWSmythe · · Score: 1

      Is running Linux or Solaris really all that different?

          My resume has a long list of Unix type operating systems on it. With all of them, I see the common features, each with its eccentricities. The same can be said of only Linux.

          Set an IP on an interface. Some want /etc/sysconfig/network-scripts/ifcfg-* . Some want an entry in /etc/rc.d/rc.inet1.conf. Some want it written directly to /etc/rc.d/rc.inet1.

          I was writing a script to get information from a couple hundred servers. In them (mostly flavors of Linux, with some AIX). To just find the OS version, I had to check for the existence of about a dozen files. Sometimes they match /etc/*release* or /etc/*version*. Some had both to provide some sort of cross compatibility, but one was a lie and one was correct. Some give it up with uname, but most give a generic line.

      root@web1:~# uname -a
      Linux web1 3.2.29 #2 SMP Mon Sep 17 14:19:22 CDT 2012 x86_64 AMD FX(tm)-8120 Eight-Core Processor AuthenticAMD GNU/Linux

      Most are pretty similar. cd changes directory. ls lists files. rm -rf / is usually a bad idea.

          Setting up a server to do common things usually has common methods.

          You can compile Apache in almost identical fashions on any of them. If you take the same version of Apache, and compile it with the same directory flags, everything will land in the same places.

          You can use a package manager to do it, but you'll have a headache of finding out what the package manager is for this kind of system, and then guessing "what directory does this distro put the conf file in?". On some you have to ask "Is it even called httpd.conf".

      (dpkg,rpm, yum, slackpkg, ipkg, opkg, pkgadd, installp, etc)

      (/usr/local/apache/conf/, /usr/local/apache2/conf/, /etc/apache/, /etc/apache2/, /etc/httpd/conf/, /etc/httpd/conf/conf.d/, /usr/pkg/etc/httpd/, /usr/local/etc/apache22/, /var/www/conf/, ""C:/Program Files/Apache Software Foundation/Apache2.2/conf", /etc/httpd/, /etc/conf.d/apache2, etc)

          Once you've worked with enough different Unix variants, you learn how to find what you're looking for. You'll know to curse various platforms for not including slocate. [finger pointing at AIX and Android], and make your own flat file of filenames (find / > ~/files.list).

          But with all those complaints, Unix is Unix is Unix, and you can use any of them once you realize that they're all almost identical.

          Mainframes are different creatures, very dependent on who the vendor was. Once you've locked in with a vendor, you're married to them for an awful long time. We have a mainframe team. They were kind enough to give me a mainframe account. I will happily admit, I don't know shit about using it. I can give them some advice on interoperability, but only from knowing the Unix side very well. I have a shell on the mainframe. I can't even attempt to do anything, but I do hope to sit down with the mainframe folks and learn some of it before the mainframe is retired and they are laid off.
         

      --
      Serious? Seriousness is well above my pay grade.
    20. Re:Deep by Lennie · · Score: 1

      AWS compute at least needs to run customer VMs, don't you think these people would like to be able to run their existing x86, euh amd64 applications ?

      Google or maybe even Facebook would be a much better example, they have their own applications with source code which they can compile for the platform of their choice.

      People currently seem more interrested in ARM processors than mainframes.

      --
      New things are always on the horizon
    21. Re:Deep by tarpitcod · · Score: 1

      What's worse is that they don't understand floating point.

      They don't understand that in floating point you can totally have a situation where:

      d = a + b + c != d= b + c + a

    22. Re:Deep by Areyoukiddingme · · Score: 1

      Undoubtedly IBM is working to "solve" that problem. By amending their contracts so they can charge for every byte transmitted into and out of their box.

    23. Re:Deep by tepples · · Score: 1

      Need a compiler for OS development work? That'll cost extra. Want the pre-compiled development API's to write applications? That'll cost more too.

      That's little different from Apple, who charges $99 to $299 per year for the right to run self-compiled applications on your own iDevice.

    24. Re:Deep by Dcnjoe60 · · Score: 1

      You can get the same precision and fault tolerance by using commodity hardware by running multiple jobs in parallel, but it's rarely required.

      It also rarely makes sense. If the parallel instances are running the same software, they will likely both make the same error, since 99.9% of reliability issues are in the software not the hardware. If you spend a million dollars on more robust hardware, and a million dollars on extra software testing (unit, integration and (especially) usability), the latter is orders of magnitude more likely to prevent a problem.

      That would only be true if you never changed the software once you spent your million dollars and tested the software. How likely is that. In reality, there is and always will be bad code out there. So, you can spend extra dollars in extra testing every time you write or change code or you can test less and rely on more costly but fault tolerant hardware.

      Face it, there is a reason that most financial instituions still use mainframes. There is also no doubt that their boards of directors want to maximize their returns. If the risk/reward tradeoff were favorable, they would switch to server farms, but so far that has not been the case.

    25. Re:Deep by Dcnjoe60 · · Score: 1

      Don't forget the low-cost dumb terminals – I'm sorry: "thin clients" – which are incapable of doing anything at all independently of the centrally-adminstered silicon. The computing environment I work in today is architecturally very similar to the one I started working in back in the mid-1980s.

      How true is that! Today's computers have so much computational power and for the most part they are being used as dumb terminals. What a waste. The PC was supposed to free us from the confines of a data center that had control of our data. Pre-internet, that looked like it was happening. Now, though, instead of advancing, we've regressed and it is 1984 all over again, except today it is a browser instead of TN3270.

    26. Re:Deep by Anonymous Coward · · Score: 1

      That's funny. I was on a project recently (ok, 1.5 years ago) comparing moving to Linux on z, Linux on p Series, and Linux on Intel. Sure, it looked pretty similar from a userspace perspective. The problem came when we considered installing any vendor software. Guess how many vendors provide binaries and will support them on Z? Hint: it's even lower than those who support P. Even some IBM products, like Websphere, didn't have complete feature parity across the CPUs. Heck, SuSE was the only distro with the same versions of packages across platforms; RedHat had a bunch of older packages on their non-intel variants, despite being the same "RHEL version", and most other distros didn't even support all three. So I'm particularly amused about a comment mentioning "providing the software depth of Linux" in the context of z Series.

      Linux on p had all the drawbacks of z with none of the benefits, so that was discarded pretty quickly. We might have kept some Linux on z, because the I/O is decent (as long as you don't need much disk; mainframe disk is *expensive*, and mainframe TCP/IP is terrible). But maintenance was a tremendous hassle. The Linux image wasn't a real disk, it was a disk image. So, if you wanted to update anything, you had to go through some other interface, update an image, and restart with that image. That makes staged roll-outs a minor challenge, and impedes package management using something like CFEngine (or one of the CFEngine-alikes like Puppet or Chef or whatever). Maybe we did that part wrong, but we did what the IBM consultant suggested. Never mind the whole cost-per-cycle thing in the mainframe licensing, and the annoying RACF integration.

      We went with virtualized Linux on Intel, and haven't regretted the decision.

    27. Re:Deep by Anonymous Coward · · Score: 1

      No layers of hardware and software error checking can counter logic errors, that's not the issue at hand at all!

    28. Re:Deep by Anonymous Coward · · Score: 0

      Care to guess how many harddrive manufacturers are competing for your business. Or how about x86/amd6 compatible processors?

    29. Re:Deep by CAIMLAS · · Score: 1

      It doesn't matter if the Lintel/Wintel/etc. machines can't do 256 DMA transfers sym.

      Chances are your vaunted NetApp or whatever Enterprise SAN storage can't do half that, either. :)

      Cool feature and all, and yes, the hardware is impressive as hell. But that's not the problem, this is:

      * Support cost (in both equipment, staffing, research, etc.)
      * Vendor dependence
      * Overall equipment cost

      Run the numbers however you like to justify it. Nobody knows mainframes anymore, not anyone with non-legacy-support resumes. And, arguably, the people who do know z are fucking idiots when it comes to anything but zOS (and so the comparision to Linux? Fairly inconsequential). You may be different, I'm not arguing against that.

      --
      ~/ssh slashdot.org ssh: connect to host slashdot.org port 22: too many beers
    30. Re:Deep by Anonymous Coward · · Score: 1

      Thanks. Too many companies have extended their useless life by doing things cheaper instead of better. (I suppose the converse is true as well). But the ability for bean counters to hide cost and inefficiencies in a spreadsheet, is met by the willingness of consumers to missapply technology and accept the resulting shit as "good enough"
      Besides the valid applications that have been weaned off mainframes, those remaining are judged by those who have never seen one.

    31. Re:Deep by Anonymous Coward · · Score: 0

      My question is why you are storing all your data on a mainframe?
      You should solve that problem before you consider what you are going to replace that hacked together crap that is Hadoop.
      The power company probably appreciates power sucking processors that spend the majority of their time monitoring hard drive failures and cores.

    32. Re:Deep by Anonymous Coward · · Score: 0

      It is nearly 100x the cost and still takes 10x longer to crunch, index and search 8PB of data on mainframe as it does in a comparatively free Hadoop cluster. The TCO was laughably different.

      Who would buy a one to 20 processor mainframe to do what a Hadoop cluster can do with hundreds of processors?

    33. Re:Deep by Anonymous Coward · · Score: 0

      At some point (and the point isn't as expensive as many people think), it gets cheaper to use a mainframe than a room full of standalone servers.

      And easier, from an administrative PoV.

    34. Re:Deep by Anonymous Coward · · Score: 0

      GP was not talking about PC's used as dumb terminals, but literally dumb terminals.

      Where I work they're getting rid of our workstations and switching us to thin client hardware -- looks to be just a monitor with USB ports on the side and a keyboard and mouse -- and a few servers running some kind of terminal services I guess. Our machines will be imaged to VM's before they're taken away, and then everything we do will be executed on shared servers, virtually, with the thing on our desk just for human I/O.

      The good news is that a workstation crash will no longer result in lost data and recovery time in a couple of days, I'm told. The bad news is if the system overall goes down (or loses connectivity), we all go down/stop working. And even when everything is working and CPU and network bandwidth are plentiful, there's still the issue of network latency.

    35. Re:Deep by baegucb · · Score: 1

      We run z/os and many Linux servers in virtual machines on a large IBM mainframe. We actually benchmarked performance on the mainframe vs. Wintel servers/clusters and found a huge performance increase on the mainframe. Otherwise, we wouldn't have switched those servers to the mainframe (heavy database usage). There are many web servers on the mainframe, but it's used for more back end work. (It could also run Windows I think I read recently, although I can't see the point). In the same room are hundreds of Wintel servers, and some AIX and HP Unix servers. The mainframe is supported by about 20 people. The servers need about 120 people by my count. TCO is lower for the mainframe, but as has been stated, it's not "sexy", so few people know the mainframe, or want to learn it.

    36. Re:Deep by dlingman · · Score: 1

      So who needs fault tolerant systems, when today, we have fault tolerant customers?

    37. Re:Deep by dlingman · · Score: 1

      Not if your servers are part of a botnet...

    38. Re:Deep by h8sg8s · · Score: 1

      Speed: Just about any 4-socket or larger x86 box can keep up with a mid-size mainframe on pure compute tasks.
      Availability: Mainframes are not even in the top tier of availability - HP NonStop is (IDC AL4).
      IO: Large UNIX systems and some others can keep up or beat mainframes at the IO game. How many MF systems have NVRAM cache?
      SW: The only reason to keep feeding a mainframe is that you have SW that *only* runs on one.

      --
      Organization? You must be joking..
    39. Re:Deep by bored · · Score: 1

      Mainframe CPU's tend to have far more error detection and correction.

      "Tend to" is now false. It has been since the nehalem based xeons. Intel seems to have picked the large scale Unix and Mainframe RAS features for the CPU and busses and made sure they can match them bullet point by bullet point.

      I challenge you to find a RAS CPU feature on your mainframe that you don't think I can find on a modern xeon.

    40. Re:Deep by bored · · Score: 2

      Because a midrange mainframe costs > a million dollars before you add the storage, maintenance etc. You want a fast mainframe your talking tens of millions.

      You can buy a lot of x86 hardware, and support talent for that kind of money.

      Personally, if wouldn't have used HADOOP. If a single mainframe can handle it, I can build an x86 server to do the same. The mainframes advantage at this point is that everything is so close to the medal it runs fairly fast. Put Java into the mix on the x86 its no surprise that it takes a cluster to match the mainframe running cobol. Especially if your trying to do IO.

    41. Re:Deep by bored · · Score: 1

      . We actually benchmarked performance on the mainframe vs. Wintel servers/clusters and found a huge performance increase on the mainframe. Otherwise, we wouldn't have switched those servers to the mainframe (heavy database usage). There are many web servers on the mainframe, but it's used for more back end work. (It could also run Windows I think I read recently, although I can't see the point).

      Wow, I would be, quite an accomplishment, what did you compare a ec12 with a bunch of single socket x86s? I have a z114 with some IFLs and a couple partitions of zos. I can show you the CPU and memory benchmarks compared to an assortment of x86 hardware. Then there is IO, apples to apples (aka match the spindles/etc) ficon sort of sucks. All the ficon disks IBM sells basically are SAS disks so your paying IO penalties against the disk subsystem. With flash disk on the PC its not even close when your talking about IOP/s.

      But what I see more frequently is someone comparing the mainframe running old fairly low level code against java on the PC. Right there your talking a huge advantage on the mainframe because java sucks for IO, and for large footprint applications the garbage collector kills cpu/memory performance.

      The fairest benchmarks I look at are linux on the IFLs vs PCs. I have yet to see the mainframe come withing 30% of mid-range x86 servers (say DL580s and IBM x3850) on any CPU or memory benchmark running C/C++ code. The IO numbers can get pretty bad because the coupling links are easily flooded if you have a decent SAN attached. The move to cabled pcie bumped the bandwidth to 80Gbit, but that really isn't anything to write home about. So, you have to be very careful about card layout in the IO expansion drawers.

      I don't have access to one of the new ec12's (maybe at some point) but the claimed 30% performance boost might put them on equal footing to some of my x86 hardware (much of which is a few years old too). I'm guessing to build a zlinux config with the same performance as my $50k DL580 would cost about 2-3 million.

      Oh, and don't get me started on zVM vs vmware. Its not really even fair.

    42. Re:Deep by Anonymous Coward · · Score: 0

      Sorry, I meant to ask the question of why would anyone buy mainframe with maximum of only, say 101 cores to do a job which has parallelism to saturate hundreds or thousands of commodity processor cores and no regulation or contract enforced time and order related constraints.. Mainframe would simply be the wrong tool for the job, no matter what the costs.

    43. Re:Deep by Anonymous Coward · · Score: 0

      It started as: "The network is the computer", and later (perhaps at a Usenix conference, I don't remember),
      "The Network is the Network, and The Computer is the Computer. Sorry for the confusion. --Sun Microsystems"

    44. Re:Deep by Anonymous Coward · · Score: 0

      RAIM memory system? Many of those RAS features are system level features anyway, so isolating CPU RAS features doesn't really do justice for either of them.

    45. Re:Deep by bored · · Score: 1

      RAIM memory system? Many of those RAS features are system level features anyway, so isolating CPU RAS features doesn't really do justice for either of them.

      Function: to assure that ECC/chipkill failures on a memory controller/DIMM/bus can be worked around.

      Its not "RAID" type functionality as such, but x86s have been doing mirroring for a long time.

      Modern ones can do either mirroring and/or sparing or variations thereof depending on the machine vendor.

      For example random google hit:

      http://h20000.www2.hp.com/bc/docs/support/SupportManual/c00502616/c00502616.pdf

      The sparing is basically as efficient (2/3 available memory) as RAIM. Mirroring doesn't use the memory as efficiently as RAIM does, but I don't see IBM publishing Read/Modify/Write timings for their RAIM systems. Probably because its _NOT_ good, doing RAID in memory is going to affect the latency of the system, which is of course probably more critical in most applications than the bandwidth.

      I have a z114 which supports this, I can probably benchmark it tomorrow with it on vs off. I don't think its turned on, but the main memory latency numbers by my measurement are pretty bad anyway so I didn't dig to deep into it.

      I suspect that IBM added the RAIM functions because they were preparing for the the flash options on the ec12. It makes sense for flash because of the issues with flash reliability and the natural page mode access of flash could be aligned with the RAIM strip size.

    46. Re:Deep by baegucb · · Score: 1

      EC12 using IFLs. Also, the SAN is the same drives for servers and the mainframe, EMC. Just the mainframe is way superior for IO. http://mainframe.typepad.com/blog/2006/03/its_the_io_stup.html Also, you can't just install vanilla SUSE Linux on a mainframe. When it was first attempted, performance was dreadfully slow. That's when IBM went, oh, you need to do some tweaks to improve performance.

    47. Re:Deep by Tastecicles · · Score: 1

      could this be part of the reason that mainframe contracts have clauses which prohibit publishing of benchmarks?

      Not that that makes much difference to me, if I'm buying a mainframe I'm not so much buying the hardware as the support that goes with it, that five-nines uptime that I don't have to be responsible for, that *I* can swing the axe if shit goes south instead of having to duck.

      --
      Operation Guillotine is in effect.
    48. Re:Deep by Tastecicles · · Score: 1

      I have three Compaq terminals, no idea where they came from or how I came into possession of them, but I know they're not cheap.

      No, I didn't pay for 'em. I'd've remembered that.

      --
      Operation Guillotine is in effect.
    49. Re:Deep by TheRaven64 · · Score: 1

      Don't forget the low-cost dumb terminals – I'm sorry: "thin clients" – which are incapable of doing anything at all independently of the centrally-adminstered silicon

      The reason that those tend to die out is that the cost of the processor is an increasingly small part of the total cost. Dumb terminals gave way to X terminals because the CRT, keyboard, I/O controllers and packaging were the majority of the cost. Adding in a frame buffer and a little display controller added a little bit to the cost, but a lot to the utility. Of course, once you'd got a display controller and RAM, adding a CPU and a hard disk didn't add very much either. If you wanted to build a thin client now, you'd stick a cheap ARM chip in as the CPU. Even something like a dual or quad-core 1GHz chip will only add $5-10 to the total cost, and adding 32GB or so of Flash also doesn't add much, at which point you have a general purpose computer. These days, the advantage of the remote service is not really to do with consolidating hardware cost, it's:

      • Better backups, because your data is on a single file server rather than scattered over your organisation.
      • Easier upgrades, because you just upgrade one machine and all of the thin clients start using the new version immediately.

      These aren't really intrinsic to cloudy services. If you have a decent file server then you can use it remotely from fat client applications and get the same storage support, and deploying applications to a whole network is something that was a solved problem even in the NetWare days.

      --
      I am TheRaven on Soylent News
    50. Re:Deep by TheRaven64 · · Score: 1

      The difference between the vertical integration of a mainframe is that the handful of hard drive suppliers and the two x86 CPU vendors are competing independently for your business. Your choice of hard drive vendor isn't limited by your choice of CPU vendor, or vice versa.

      --
      I am TheRaven on Soylent News
    51. Re:Deep by tehcyder · · Score: 1

      rm -rf / is usually a bad idea

      Luckily there is no command line in Windows so I can't make that sort of

      --
      To have a right to do a thing is not at all the same as to be right in doing it
    52. Re:Deep by bored · · Score: 1

      For some reason I can't find the paper linked in that article. But accessing 64k devices is basically nothing, especially if they are a bunch of virtual mod3s and mod9's like just about every mainframe config I've seen. And really its less than 64k because your terminals, tapes, etc are also in that address space. Its like saying I have 4 Gigabytes of IO space and each adapter only takes 16 bytes of it. Useless BS information. Really, I was thinking about the address space issues on the mainframe last year, and came to exactly the opposite conclusion. Its a limitation, especially since the address space is shared between LPARs. AKA if you have 20 LPARs, with independent devices its actually only 4k devices per LPAR, which could be really limiting, especially if you can't run EAVs and the max capacity per volume is 54GB.

      But the real problem is that even if you attached 256 devices on multiple 8Gbit ficon channels, the CEC to IO drawer interconnect is going to be the bottleneck. Thats assuming you can start enough ficon commands to keep each channel busy.

      Here are real numbers from IBM on their ficon performance at 8Gbit (still current speed).

      ftp://public.dhe.ibm.com/common/ssi/ecm/en/zsw03127usen/ZSW03127USEN.PDF

      Go look at emulex's site. Their numbers are over an order of magnitude better and they offer a 16Gbit board too.

      Anyway, did you ask EMC why the numbers were worse on your other platform? I'm betting you can "tweak" it up to speed too. Rarely is the machine the bottleneck when talking on a SAN. Its almost always the disk subsystem. Linux out of the box on nearly every platform needs "tweakage" Many of the fiber channel drivers on linux have very shallow queues and need to be significantly deeper for IOP benchmarks.

    53. Re:Deep by ebno-10db · · Score: 1

      Install Cygwin.

    54. Re:Deep by Anonymous Coward · · Score: 0

      I disagree The capability of an aggregation of servers differs from a mainframe, in it's ability to share work. Some parallel tasks are embarrassingly parallel (http://en.wikipedia.org/wiki/Embarassingly_parallel) but most are massively parallel (http://en.wikipedia.org/wiki/Massively_parallel) at best. That's why all the computers on the internet working together, don't amount to a single really huge supercomputer.

    55. Re:Deep by ArsonSmith · · Score: 1

      I'm not, customers are.

      --
      Paying taxes to buy civilization is like paying a hooker to buy love.
    56. Re: Deep by Anonymous Coward · · Score: 0

      IBM mainframe has cheap commodity MPP co-processor option Transparantly tightly-coupled, which z/os DB2 exploits unknown to ad-hoc query BI. IBM mainframe now leverages Netezza's solution-set having MPP FPGAs cherry-picking rows/columns and doing aggregation effectively as disk-controllers with lower carbon and floorspace footprint ( lower A/C and lower power)...and FPGA tech is improving at rate faster than Moore's law projected to be accrued over longer period. I wish I had mainframe "SMF" (session management facility) on Linux to be logging every event for ETL file interarrival rates or message-processing guaranteed store-and-forward. Mainframes can and do run Java, often better than midrange, but most often in parallel with midrange. SQL is stored in JARs inside DB2 itself, transparent to Java Luddites, moving process to the data instead of data to the process in spite of Java Luddite efforts to the contrary. Java Luddites enjoy their single-processor funnel slugging along in a century where multicore is the norm, where CPUs sit idle if data mist move from more than 2cm away. MPP boxes running RDBMS have exploited dual-core for years, yet Java Luddites still insist that single-threaded OOP systems are high-tech, while hide-tech realities fail to handle cross-cutting concerns such as logging and security that don't fit in OOP paradigm. Earlier post illustrated server consolidation of Linux/Unix boxes are nearly transparent today for RAS(reliability, Availability, Serviceability) factors, as well as TCO(total cost of ownership studies weighing engineering economics present-valued impact assessments(including present-valued cost/benefit/ROI/IRR weighted with goal-alignment and risk-assessments all included) at least within companies already having mainframe infrastructure management Talent pools,. Interestingly, just as the world passed by the old COBOL mainframe Luddites in the last millennium, the old dinosaur/Phoenix Mainframe is presenting next-generation paradigm shift alternatives where exploiting tightly-coupled MPP RDBMS such as Netezza with latest FPGA and even multi-core exploitation might be a more viable solution than trying to get Java to more rapidly exploit multiple-core tech, leveraging tools that transparently unravel Hibernate and RDBmS statements, convert them to "static C" programs and run "inside DB2" as JARs, with feature-rich full bi-temporal and context-level security and pure XML support (even full PL/SQL) support oftentimes not even known nor understood never mind exploited and leveraged by Java Luddites, who are happy reinventing the single-core wheel of bi-temporal and XACTML rules-engines and cross-cutting concerns instead of using feature-rich solution set bought and paid for capabilities engineered-in already into RDBMS capabilities who would rather re-invent in support of the 2013 IT developer full-employment-act, whilst begging for java compilers to save-them, save-them, save-them from multi-core oblivian. Trying to figure out how to make paradigm shift to functional programming to begin to exploit multi-core whilst resisting the fact that RDBMS functions and stored-procedures and queue-tables and FPGA does it for them and unravels their OOP spaghetti-code for them unbeknownst to them to run as efficient JARs inside MPP engines in spite of their optimum single-core design and Java Luddite efforts. The tech moved on, back to mainframes and z/Linux and z/os and Netezza and ziiPs and ZaaPs and even z/hadoop with full SQL support, when they were not watching....pity that the pendulum swings both ways...back and forth...built in obsolescence in IT workforce Luddite lands! ;^D

    57. Re:Deep by baegucb · · Score: 1

      I presume this comment reply will get emailed to you. Mine are, but it's a seldom used email box of mine. Anywats, we use 8 IFLs and despite what wikipedia says, our in house hardware experts say they are faster then general ones. And cost less.
      As far as is it more cost effective? Trust me when I say our organization was under immense pressure to cut costs. Our customers are very happy.
      Comments will be cut off on this article, and I'm on vacation. But my nick at gmail, and you can email me. Meantime, I'll copy/paste to my work email and give you a more detailed answer if you wish. I'm currently sorta the z/os side, but AIX and HP-UX i the past.

  2. Running Virtualization Software.... by Anonymous Coward · · Score: 4, Insightful
    1. Re:Running Virtualization Software.... by Anonymous Coward · · Score: 0

      Yes ... virtualization.

      The only difference is an architectural one.

      A mainframe is "one" entity that could run many virtual environments. As more you need to run, the mainframe must be more powerful.

      A virtualization server (today usually we must talk about x86 based ones), is a powerful machine that runs a Hypervisor to host many virtual machines.

      A computer room with many virtualization servers accomplish the same role as the mainframe, although technically is different, because it is not only one machine; however, management products are trying to make everything to look as one big device, making the modern and the old styles to behave in the same way.

      Could be possible that these management/hypervisor products will evolve to become the next VM?

    2. Re:Running Virtualization Software.... by Anonymous Coward · · Score: 0

      It's worth remembering that the biggest mainframe seller, IBM, hasn't been competitive in the processor market for ages. They can't make anything to compete with Intel, AMD etc..

      It's also worth remember that they don't make disks anymore, they just buy them in.

      They also don't compete in the OS market, their software isn't competitive.

      So on the one hand, the server maker, uses the fastest processor, the fastest ram, the fastest disks. On the other IBM uses their own 3rd or 4th tier products to make a mainframe.

      Which do you think is going to be faster? The IBM Mainframe or the PC Server?

      Nobody should be paying premium prices for 3rd tier kit these days, (legacy software problem apart).

    3. Re:Running Virtualization Software.... by NighthawkFoo · · Score: 1

      IBM fabs its own CPUs for the mainframes, and they are currently clocked at 5.5GHz, which is the fastest shipping processor in the world.

      https://en.wikipedia.org/wiki/IBM_zEC12_(microprocessor)

      --
      "I disapprove of what you say, but I will defend to the death your right to say it."
      - Evelyn Beatrice Hall
    4. Re:Running Virtualization Software.... by bored · · Score: 1

      Fastest clocked != best performer..

      IBM basically doesn't have a power budget for these things (although they claim the new ones are greener). That allows them to build machines with blower motors that are two feet across and heatsinks that weigh more than my whole desktop.

      That doesn't mean they run super fast, Intel and IBM with POWER (and others) have done amazing things with the power budgets they stay within.

    5. Re:Running Virtualization Software.... by TheRaven64 · · Score: 1

      It also depends a huge amount on the workload. For example the latest Z series and POWER chips have a reasonable amount of die area dedicated to floating point binary coded decimal. For most code, this is wasted space. For COBOL applications, it can be an order of magnitude speedup.

      --
      I am TheRaven on Soylent News
    6. Re:Running Virtualization Software.... by NighthawkFoo · · Score: 1

      They also support a liquid cooling option :)

      --
      "I disapprove of what you say, but I will defend to the death your right to say it."
      - Evelyn Beatrice Hall
  3. Privacy by MLBs · · Score: 5, Insightful

    It's the usual argument. If you have something to hide, you're probably a bad person.
    That "may" be true if the authorities are not abusing their power, or trying to gain more power than the people want them to have.
    As soon as you have even a potentially oppressive regime, privacy becomes essential.

    1. Re:Privacy by Anonymous Coward · · Score: 0

      I'm starting to accept that there is very little privacy left. At first it was very upsetting, but I'm slowly becoming indifferent to it, there's not much I can do about. Maybe an upside to all this will be that people will try to behave better because they know they might be shamed by their actions. Kind of similar to the days of old in a small town/settlement.
       

    2. Re:Privacy by Anonymous Coward · · Score: 0

      As I post AC...

    3. Re:Privacy by Anonymous Coward · · Score: 0

      I looked at your "people will try to behave better" through this pair of shades, and for some reason it reads "CONFORM OBEY"

      Strange...

    4. Re:Privacy by kheldan · · Score: 1

      That "may" be true if the authorities are not abusing their power

      Oh, they are. It's just more subtle, I think, than in past attempts. The rise of "social media" like Facebook has indoctrinated an entire generation that "sharing is good and healthy, hiding (meaning: privacy) is bad and unhealthy" and you get laughed at for wanting privacy at best, dirty looks and attacked/accused of outrageous things at worst. The sad part is they don't realize what it is they've given up until it's too late to do anything about it. I'd have a hard time believing that there are not people in positions of power who are quietly and subtly encouraging the anti-privacy attitude. There do appear to still be some people in positions of power who are still working to preserve citizens' privacy, but who knows if it's enough?

      --
      Are YOU using the TOOL, or is the TOOL using YOU? Think about it!
    5. Re:Privacy by Anonymous Coward · · Score: 1

      A major problem is asymmetry. You have nothing to hide if you are doing nothing wrong, yet, the people in power get to hide everything. If EVERYONE could access everyone else's information, then I am okay with that. The fact that a privileged few access the information of all others is the thing that bothers most people. Therefore, either all people access information, or privacy is essential.

    6. Re:Privacy by Voline · · Score: 2

      I think you're misreading the article. The Winestock is not making the "if you have something to hide ..." argument, he's anticipating it. His argument is that the computer industry, and perhaps computing as a technical endeavor, tends the direction of centralization of computing power and grunt work which then leads to centralization of data. Both governments and business – even cool, supposedly "revolutionary" businesses – like it this way. So, don't look to the high tech companies for help protecting your privacy. As he says in TFA:

      Pleading will not help because the interests of those companies and their users are misaligned. One reason why they are misaligned is because one side has all of the crunch; terabytes of data, sitting in the servers, begging to be monetized. Rather than giving idealistic hackers the means to liberate the users from authority, the democratization of computing has only made it easier for idealistic hackers to get into this conflict of interest. That means that more of them will actually do so and in more than one company.

      You see, in the past, the computer industry was dominated by single corporations; first IBM, then Microsoft. Being lone entities, their dominance invited opposition. Anti-trust suits of varying (lack of) effectiveness were filed against them. In the present, we don't even have that thin reed. Thanks to progress, we now have an entire social class of people who have an incentive to be rent-seekers sitting on our data.

      Being members of the same social class, they will have interests in common, whatever their rivalries. Those common interests will lead to cooperation in matters that conflict with the interests of their users. For example, the Cyber Intelligence Sharing and Protection Act (CISPA) is backed by Microsoft, Facebook, Yahoo, and, yes, Google, too.

      As the head of the Software Freedom Law foundation, Eben Moglen says, keep your data locally, at home, where the 4th Amendment still has some effect. As Winestock is saying, you better be ready to defend even the right to do that.

    7. Re:Privacy by Anonymous Coward · · Score: 0

      From the sound of it, it's more likely you came to realize that to achieve further levels of Progress, privacy and other freedoms must be phased out.

    8. Re:Privacy by BrokenHalo · · Score: 1

      in the past, the computer industry was dominated by single corporations; first IBM, then Microsoft. Being lone entities, their dominance invited opposition.

      And some of us did quite nicely out of that at the time. I was one of a number of sysprogs who made a pretty good living out of working contracts on CDC, Burroughs, Honeywell and Sperry mainframes. Back in the '70s and '80s there was very much an "us and them" mentality: those whose idea was that a computer meant IBM, and us who built up massive CVs working across the field of everything else. I personally found the latter much more exciting. Even as the huge power-guzzlers of the '70s and '80s gave way to the more compact Prime, Data General and Unisys machines in the early '90s you still had to work hard to keep your reputation in the field.

      Back then, of course, your data mostly was stored offline. But probably 90% the time, a tape-monkey^W^W computer operator in a larger computer room would never recognise a message from a dubious PID on his console for a rollback tape, and there goes your security.

      Mainframe computers have acquired something of a reputation for security, but nevertheless, during the Cold War there was a widespead (and justified) nervousness about data leaking to the wrong people.

  4. Having worked in both by mbone · · Score: 3, Interesting

    He is wrong, on pretty much every level, even the visual.

  5. Well well, by Fuzzums · · Score: 0

    "If you have something that you don't want anyone to know, maybe you shouldn't be doing it in the first place."

    Ok. First of all, I said [xxxxxxx] Krist. Second, I'm proud of my Willy and I want everybody to know. Not only that. I want to show it on television and I don't want Krist extremists or parent extremists to censor me. I have a Willy, You don't respect my privacy, then I have to confront you with my Willy. Simple as that.
    Also I want you to know I went to the toilet. I want you to know that too. I mean, no. I don't want you to know, but I did it and you said you wanted to know.

    You see. That privacy thing. It has a reason. There is A LOT I did do that I don't want you to know. Just because it's not of your fucking business.

    And now I'm going to [xxxxxxxxxx].

    --
    Privacy is terrorism.
    1. Re:Well well, by tedgyz · · Score: 2

      Also I want you to know I went to the toilet. I want you to know that too. I mean, no. I don't want you to know, but I did it and you said you wanted to know.

      Sounds like a typical facebook post. People are giving this information away willingly. Some of us want privacy, while others want to tell us every last detail of their lives. I would like to not have to read about peoples every move, which is why I unfriend those that share too much.

      --
      "No matter where you go, there you are." -- Buckaroo Banzai
    2. Re:Well well, by Fuzzums · · Score: 1

      I'm scared! I don't want to go to G'bay because "they" might think I value my privacy. So I HAVE to share.

      --
      Privacy is terrorism.
    3. Re:Well well, by Anonymous Coward · · Score: 0

      Ah, but see, no where in that quote does it suggest that the thing you are doing is wrong. Only that, in this day and age, if you don't want something getting out, just about the only method of ensuring that is to not do that thing.

      Think of it like Abstinence Only Sex Ed.

  6. Ending maintenance also ends control by h2oliu · · Score: 5, Insightful

    One of the points I found the most insightful is that the geeks don't like to take the time to make things work anymore. I remember a colleague saying that there was no better way to kill a hobby than to get it as a job.

    The days of tweaking the OS and hardware as a common practice among the majority of geeks is gone. The field is too broad now. You have to pick which stack, and where on it, you want to hack.

    --
    Ok, I give up, why you?
    1. Re:Ending maintenance also ends control by tarpitcod · · Score: 5, Interesting

      Back in the earlier days of micros it was loads of fun. BYTE was a great read. People wrote their own stuff on their own hardware. There were really fascinating choices in CPU's. Initially there were people using 2650's 8080's, 6502's, 6800's, LSI-11's, 1802's, 9900's. .

      I can't remember the last time when someone actually said something outrageous like 'What architecture would be ideal'. Nowadays it's 'What software layer (implicitly running on x86 Linux boxes) should we use?'

      The performance numbers people talk about are terrible too. Kids who just graduated think 100K interrupts per second is 'good!' on a multi Ghz multicore processor. They just have no context and don't understand how absolutely crappy that is and that even on an 8031 running at 11 Mhz with a /12 clock we could pull off > 20K interrupts per second in an ISR written in HLL!

    2. Re:Ending maintenance also ends control by rastilin · · Score: 1

      That sounds too simple, there must be a reason for it other than the new youngsters suck at programming compared to the older generation.

      --
      How do you kill that which has no life?
    3. Re:Ending maintenance also ends control by mikael · · Score: 3, Informative

      And if you know where to look, you can find the whole collection of magazines scanned and available online (http://atariage.com/forums/topic/167235-byte-magazine/)
      The best issues where when they had geek cartoons or photographs of real hardware on the front cover. The real change was when everything went all pastel shaded with the little bod characters in suits. I guess that coincided with the shift from hardware projects to software API programming on personal computers.

      --
      Vintage computer adverts: http://www.vintageadbrowser.com/computers-and-software-ads
    4. Re:Ending maintenance also ends control by tarpitcod · · Score: 1

      It's great to see these things still around. They are really fun to read. I actually am a bit of an atari 8 bit fan with some 8 bits I still use occasionally for fun.

    5. Re:Ending maintenance also ends control by tarpitcod · · Score: 4, Interesting

      Try finding out yourself. Ask some kids some simple questions to the new kids:

      Try asking them:
      What's the memory bandwidth of that x86 desktop or laptop roughly? Special points if they break out cache.
      Ask them how many dhrystone MIPS (very roughly) that uP has.
      Ask them the ratio of the main system memory bandwidth to MIPS.
      Ask them the ratio of the main system memory bandwidth to the I/O storage they have.

      They just never get exposed to this stuff. They just have no reference. Now ask them to compare them even to a regular 286 era ISA bus PC: I'll even give you some numbers.

      286/16 ~ 4K dhrystone MIPS on a good day
      Disk (40 MB IDE on ISA) ~ 400K/sec

    6. Re:Ending maintenance also ends control by Anonymous Coward · · Score: 0

      That spirit is alive and well in the high-end graphics community, though even that is getting fairly standardized. Console hardware usually has some very interesting kinks in it that allows someone who knows what their doing to really put the squeeze on existing hardware. (Which is why games look better even 6-7 years into a generation - the tricks really help you squeeze out the drops of performance).

      You won't find that spirit in "web stack" programming. If you're already in an interpreted language, you don't really care about the hardware. Its all virtual anyways, you just need "20 medium instances" and "10 large ones for the heavy lifting". Which is fine I guess, but there are sometimes when the swiss army knife of generic servers won't do the job. Take for instance SSDs. One different server outfitted with SSDs will crush anything else as a logging/graphing box, yet the guys building out the servers (or rather outsourcing the buildouts) look at you like you have 3 heads when you make the request.

      All that said, I'd probably still prefer growing up today in computers rather than the 70's/80's. For one the internet can answer your questions so much easier, where then it was maybe a book at the library, more often than not something special order, and a little later on BBS forums and stuff. It was all pretty "small". Today's programming environments and packages are so much easier to so much more with just as much investment in time. Sure you don't spend time getting the TCP stack running, but that time investment is now spent on building a server or something more interesting.

    7. Re:Ending maintenance also ends control by drinkypoo · · Score: 1

      The days of tweaking the OS and hardware as a common practice among the majority of geeks is gone. The field is too broad now. You have to pick which stack, and where on it, you want to hack.

      I tried to git clone the android-x86 repo per the instructions and it just never completed and kept dying, so I wound up with a 21 GB .git directory and nothing else. The people who have the bandwidth are buying new hardware and expect it to just work. The people with the old hardware don't have the bandwidth. Back "in the day" your whole OS and all the sources would fit on a stack of floppies or on one CD with room left over and you could reasonably download a new OS via POTS MODEM. Today, I literally cannot fetch the sources and actually even make a contribution to some systems. I do have a rinky-dink connection, but that's not changing...

      --
      "You're right," Fisheye says. "I should have set it on 'whip' or 'chop.'"
    8. Re:Ending maintenance also ends control by CAIMLAS · · Score: 2

      The reasons why people don't 'hack' their stuff anymore is because:

      * They're working 60+ hour weeks and don't have the time
      * The people who used to are now adults, with responsibilities outside of work, and don't have the time
      * Kids these days aren't really all that interested, unless we're talking about mobile handsets (aka smartphones), which DO get 'hacked' a lot.
      * There's usually no point in making small scale changes. Shit is fast enough now; you don't see a 20% increase in performance by 'tweaking' things, usually - like you used to.
      * People who are still hacking things have server closets (whatever) at their house and have learned how to do a lot more with their time than 'hacking' settings would allow for (allowing them to be more productive at work).

      --
      ~/ssh slashdot.org ssh: connect to host slashdot.org port 22: too many beers
    9. Re:Ending maintenance also ends control by VortexCortex · · Score: 2

      And their answer to every one of your questions is: WAY more that I even need! And this is a USED box!

    10. Re:Ending maintenance also ends control by Darinbob · · Score: 1

      I see a cycle going on too. First mainframes. Then because that was a pain to tinker with and you had no control, departments would buy their own minicomputers. Over time, that got to be a lot of hassle plus the main computer organization wanted to keep control, so the ownership of the minis ended up leaving the departments and becoming centralized all over again. This repeated with the workstations as well. Individual departments or even individual workers would get a workstation to do their own work independent of any centralized priesthood. Useful things but a lot of researchers found that managing and maintaining them got into the way of doing actual research, and again these workstations slowly reverted to being centrally controlled. Then the PC, repeat all over again, cheap enough to be on the budget of a lot of departments and workers, free of any official oversite, with some useful applications that were clumsy on the central computers, so you just set one up next to your mainframe terminal. Again, the individual workers discover that it's no fun being a hacker type and maintaining your own PC or mac, plus the internet shows up along with viruses, and the centralized priesthood demands full and absolute control over any electronic device in the office.

      Overall you start with some people who don't mind tinkering with computers and who want their own, and they are using the computers for specialized operations that aren't handled well by the centralized computer organization. Then over time once these become established the number of users who are willing to tinker with and maintain them gets smaller. Control is given up but you still have some workers who have some use for the individual control or thinking outside the box. Then those workers repeat the cycle again.

      All the while, the same old centralized machine room has existed with little change. First one giant mainframe, then a few large minis, then a lot of racks for NFS and related servers, then racks of PC servers and network routers and databases.

    11. Re:Ending maintenance also ends control by gman003 · · Score: 2

      As one of those damn kids, let me try:
      Let's see... it's a dual-channel DDR3 memory controller, so that's 128 bits per transfer. DDR, so two transfers per clock. And clocked at 1333MHz, so 341,248 Mb/s, or 42,656 MB/s (I'll call it 42GB/s for short). Cache I'd have to look up, but I think L1 and L2 caches are synchronous to CPU clock, while L3 is running at half-clock. L1 I think reads in 256-bit cache lines, not sure about any of the others.

      I personally have never needed to use dhrystone - I'm one of those people who dislikes synthetic benchmarks, just on principle, particularly ones that ignore floating-point operations completely. Looking at numbers others have gotten, I'd estimate this laptop to be in the 7000-9000dMIPS range. I can, however, calculate the theoretical instruction rate at around 25GIPS, although in any practical use I would expect closer to 15GIPS, even if purely CPU-bound and using only single-clock instructions.

      Using that 15GIPS figure, that gives a ratio of memory bandwidth to instruction rate of roughly 2.8 bytes per instruction. That seems pretty reasonable, especially when you consider that, in a memory-bound workflow, the CPU would not be overclocking itself so much, improving that ratio.

      Storage will get a bit complex, since I have both an SSD and a hard drive in this machine. The SSD runs around 350MB/s on sequential read, and around 250MB/s on sequential write, which gives a ratio of either 120x or 170x depending on whether you're reading or writing. Random I/O will also work differently, but not significantly for any "random" workload I'm likely to run on this thing. The disk runs about 180MB/s on sequential work (even though it's a laptop drive, it's still 7200RPM), which makes the RAM about 230x faster (of course, random access would quickly turn that into a four-digit number).

    12. Re:Ending maintenance also ends control by Anonymous Coward · · Score: 0

      You're trying to fetch the entire source history of android-x86. Operating systems on distribution media are usually binary-only and optimized for size.

    13. Re:Ending maintenance also ends control by tarpitcod · · Score: 1

      Good on you for doing the exercise! I agree dhrystone isn't perfect but I wanted a benchmark thats relatively easy to find numbers for a machine.

      For the 286 there's about a 10:1 ratio. Your ratio is 230:1. So let's just scale it and see what that would have looked like: 4000 dhrystone mips / 230 ~= 17 K/sec.

      So effectively your I/O to compute ratio of your laptop is about that of a 286/16 with a floppy disk.

      The bad news is it gets *worse* than your laptop with typical x86 servers. I/O doesn't scale like compute. Take a x86 box people might throw VM's on. It's probably got worse I/O versus compute than your laptop or a 286/16 with a floppy.

      Totally right about latency too. That's an even more depressing ratio.

    14. Re:Ending maintenance also ends control by walshy007 · · Score: 1

      You're off on the memory bandwidth by a fair chunk, check this out.

      Haven't checked your other numbers though. On my machine with dual channel ddr-1333 I get about 16gb/sec, and l1 cache is about 60gb/sec

    15. Re:Ending maintenance also ends control by Tablizer · · Score: 1

      ...and we used to kick people off our lawns by hand, not via hacked Roomba's; that's for cowards!

    16. Re:Ending maintenance also ends control by StoneyMahoney · · Score: 1

      For the overwhelming majority of software development graduates over the last decade, these figures are completely irrelevant. Only a tiny fraction of them will ever actually compile something into native code that will be run as the sole task of an entire CPU. With modern hardware load-optimization techniques, especially things like Intel's Turbo Boost, even a single CPU can be a dynamic target during a single run - depending on what else the system is doing at the time.

    17. Re:Ending maintenance also ends control by gman003 · · Score: 1

      I just forgot that the "1333MHz" figure was actually "1333MT/s", already having the "two transfers/clock" multiplied into it. Thus giving me twice the bandwidth I actually have.

    18. Re:Ending maintenance also ends control by Anonymous Coward · · Score: 0

      Hardware is also stupid cheap now. A 20 dollar raspberry pi is more capable that a 2000 dollar 386 from 1990.

      Our hardware is head and shoulders above what the old tech could do.

    19. Re:Ending maintenance also ends control by RockDoctor · · Score: 1

      the geeks don't like to take the time to make things work anymore

      Sorry, but geeks do take the time to make things work.

      A lot of people who call themselves "geeks" don't take the time to make things work, but that's because they're not geeks, just people who call themselves geeks. Similarly, I can go to any random lunatic asylum and round up a herd of people who call themselves "God" (or trivial variants of that), but the only way to tell if they really are God (or a close approximation) is to throw them in the lake and see if they can walk out without getting their feet wet.

      I remember a colleague saying that there was no better way to kill a hobby than to get it as a job.

      I remember having the same conversations as a child with family, friends and teachers - mostly people wiser than me at that time. And so, you know, I actually followed their advice and do something for a living which is not my hobby. (Or not my main hobby, at least.)

      There is this thing about advice ... if you don't listen to it and don't follow it, that's your responsibility, not the responsibility of the people who gave you the advice that you've asked for.

      --
      Birds are not dinosaur descendants;birds are dinosaurs, for all useful meanings of "birds", "are" and "dinosaurs"
  7. Mainframes is for those.. by i · · Score: 4, Insightful

    ..that have very big amounts of data, complex data structures and can't afford any errors (especially data corruption) caused by hardware limitations.

    Banks is an example.

    --
    Mundus Vult Decipi
    1. Re:Mainframes is for those.. by Nerdfest · · Score: 1

      Big data is more readily done with racks of commodity hardware. You get orders of magnitude better performance for the money. Do you seee any of the big web companies moving to mainframes? If there was cost or performance improvements in it they'd have done it in a second.

    2. Re:Mainframes is for those.. by jonwil · · Score: 3, Insightful

      The people who use mainframes for big data (like banks and insurance companies) and the people who use clusters and racks of servers for big data (like search engines, social networking sites and other web companies) have totally different requirements.

    3. Re:Mainframes is for those.. by fermion · · Score: 2
      I have seen large complex data sets on racks of cheap microcomputers in places wehre i work. We see this in Google, for example. What characterizes these data sets is that are easily replicated, or there is little liability if there is loss. Think about data loss on google and then think about a bank misplacing a deposit. Do we think that Google keeps many of it's algorithms secret for no reason? No, they do it so they are not held accountable.

      For servers facing the internet, load balancers, like those made by compaq in the late 90's, do a very good job treating those servers as a RAID. The question is how much is the data changing behind the servers, the liability is data is compromised, and as mentioned how complex managing the data is. The question is also how complex it is to manage a hundred thousand machines instead of one big machine.

      --
      "She's a scientist and a lesbian. She's not going to let it slide." Orphan Black
    4. Re:Mainframes is for those.. by Type44Q · · Score: 1

      And apparently are is quite dead... :/

    5. Re:Mainframes is for those.. by meta-monkey · · Score: 4, Interesting

      I think you're confusing Big Data with big, data-reliant companies.

      Banks are OLTP, and require perfect accuracy, [large number] 9s uptime, fast response, dealing with one record at a time.

      Big Data is OLAP, and can sacrifice some speed, accuracy and uptime to operate over millions and millions of records.

      --
      We don't have a state-run media we have a media-run state.
    6. Re:Mainframes is for those.. by Anonymous Coward · · Score: 0

      Banks is an example.

      Banks ARE an example.

    7. Re: Mainframes is for those.. by Anonymous Coward · · Score: 0

      Nope, Banks is the example. He's the only guy in this shop who understands JCL. Would be nice if he showed up to actually do his job once in awhile....

    8. Re:Mainframes is for those.. by Anonymous Coward · · Score: 0

      But it's only one example. It's singular, therefore is.

  8. Let's bring that paranoia out front and center! by __aaltlg1547 · · Score: 1

    We all need a good look at it. Does it look ridiculous to everybody? Good. Now let's move on to things that might actually happen.

    Server farms will offload much of the computer power and most people will use lightweight, low power portable devices? Yeah probably.
    Server farms will get bigger and more powerful? Definitely.
    That model will fit for every business and organization and individual user? No way. Won't happen.

    Please keep in mind that my 3 year old Android phone is more powerful than any PC was in 1990.

    1. Re:Let's bring that paranoia out front and center! by Anonymous Coward · · Score: 1

      > Please keep in mind that my 3 year old Android phone is more powerful than any PC was in 1990.

      And the really sad thing is, it feels slower than that slow PC did back in 1990.

      I remember running Windows 95 on a grossly resource-starved Compaq Aero 33c back in 1996. It was so primitive, it could only do 16-color VGA, and had to bitbang audio by direct speaker-manipulation. The whole hard drive was compressed with Microsoft's version of Stacker, and it had a 170mb hard drive with 12mb of ram and a 486/33 CPU. Somehow, it managed to run WordPerfect for Windows well enough to use, though I usually ended up using WP5.1 for DOS just because WPwin crashed so much.

      Fast forward 3 years, to Windows 98 SE on a Pentium 100 laptop with 32 megs (Toshiba Satellite, I think it was a CDS 400 series). Utterly and completely fucking unusable, courtesy of Active Desktop, until I used 98Lite to rip out AD and replace it with Windows 95's version of Explorer. At one point (before 98Lite), I was actually navigating the mouse pointer around the desktop and being careful to avoid accidentally passing over anything, because grazing any icon would make the whole computer grind to a halt until Windows finished building the objects and fetching the metadata for it. Had it been a more modern cloud-based OS with forced auto-updates, ripping out AD wouldn't have been an option.

      Now, skip over the happy island of relative stability that constituted Windows 2000 (up until ~SP4, at least), and use XP... where everything is somehow bound to Internet Explorer or dependent upon it for something. Try to load some malfunctioning site on the internet, and watch everything else on the computer grind to a halt while the recursive DNS resolution is taking place because Microsoft stupidly made it a single-threaded chokepoint that everything has to pass through, even if the resolution job in the queue is "c:\"

      Fast forward a little more than a decade, to the abomination that constitutes my work computer. It has a pair of 4-core i7 processors, 16 gigs of ram, a SSD, and -- thanks to IT policy -- is determined to store just about everything on network shares, so anything that involves file i/o (and quite a few things that don't, at least not overtly) makes the computer stall and stagger as badly as my underpowered P100 laptop did 15 years ago.

      It's a nightmare. Click on an email in Outlook, and wait... and wait... and about 7-20 seconds later, it finally opens (probably after dimming the window to indicate that, as far as Windows is concerned, the program is about to crash). Double-click a Word document, and go visit the restroom while it's loading (a document that would load in about a second on my computer at home). Then, when you're feeling REALLY masochistic, open IE and launch one of the company's browser-based applications. Click... and wait. And wait. And wait. Or worse, go to click something, and have the screen re-compose itself a fraction of a second before you click, and end up clicking the wrong thing. Then wait, wait, and wait more until you can finally back out of it and get back to where you were.

      The point is that over time, our hardware has gotten faster, but our computers still manage feel slower and slower. Our computers are now hamstrung by network delays. Server-based apps are worse, and apps written in interpreted languages like Javascript that are hosted in browser containers with massive network-based dependencies are the most toxic of all.

      Say what you like about Java... at least it has JIT compilation, and runs well (on Windows, at least) once it's up and running.There are few things I despise more than Ajax, keystroke-by-keystroke network calls, Javascript, and browser-based applications in general. Ten years ago, web-based apps were slow because the network was slow. Then, we got faster connectivity, but the servers were slow because they were overloaded. Then the servers got upgraded, but they replaced one burst of network data with a hundred thousand microbursts that never end, and

    2. Re:Let's bring that paranoia out front and center! by Jonner · · Score: 1

      Please keep in mind that my 3 year old Android phone is more powerful than any PC was in 1990.

      How much can you do with that computer in your pocket without depending on "the cloud"?

    3. Re:Let's bring that paranoia out front and center! by __aaltlg1547 · · Score: 1

      I can read my email off line, open and edit documents, etc and quite a number of other things. I find the poke-with-my-fingers input and tiny screen UI to be more of an issue than the cloud dependence. I also have direct access to some of my other computers (those I have set up to allow me to log into them from my phone). The cloud dependence is not much of an issue because it's -- you know -- a phone. The reason I even HAVE it is for mobile connectivity. The reason I have the other computers is more mixed.

  9. Datacenters === Today's Infrastructure by Anonymous Coward · · Score: 0

    Today's datacenters where of course not needed in the past as they are now.
    The quantity, and the individual size, of the data transmitted is growing larger with new users and media.

    Are the majority of YouTube videos necessary?

  10. Mainframes and server farms the same? Hardly by div_2n · · Score: 2, Informative

    I suppose if you stand back from about 3 miles and never bother to understand the underlying architcture and how it scales while ignoring the flexibility of server farms as opposed to very much a box that mainframes put you in (with very minor flexibility) then yeah -- they're exactly the same.

    It's easy to draw parallels between general functionality, but you have to reduce it to "a series of tubes" type descriptions to get there.

    1. Re:Mainframes and server farms the same? Hardly by Lennie · · Score: 1
      --
      New things are always on the horizon
    2. Re:Mainframes and server farms the same? Hardly by nanoflower · · Score: 1

      I didn't get the same meaning as you did. My reading is that the mainframe and server farms are the same in that they centralize information. Giving the corporations access to and control of much of your personal data. That's something that we had begun to move away from with the rise of the personal computer but the move to the cloud is going back in the other direction. I don't think that's going to stop because the cloud and server farms provide the user some great benefits but it's worthwhile to keep in mind just how much information corporations have on you. Especially with the bills underway to give the government easier access to all of that information when ever they want.

    3. Re:Mainframes and server farms the same? Hardly by VortexCortex · · Score: 1

      I suppose if you stand back from about 3 miles and never bother to understand the underlying architcture and how it scales while ignoring the flexibility of server farms as opposed to very much a box that mainframes put you in (with very minor flexibility) then yeah -- they're exactly the same.

      I think that was his point exactly:

      "In his latest essay, Rudolf Winestock argues he's about as bright as a box of rocks and that the movement to replace him with a bag of mostly water is simply re-inventing the dumbness, and will result in the same level of advice they were being given in the first place. Also, it's a conspiracy."

      You fail to see through to the real message hidden between the lines, which is: It doesn't matter what you say if you're talking to idiots, so long as you're slightly more knowledgeable than them and the sounds you make are vaguely recognizable, they'll probably believe you, especially if you get some other idiots to fall for it first then cite you by name.

      I personally believe the submission is a call to action about Wikipedia, but with this level of performance art you never can tell, truly.

    4. Re:Mainframes and server farms the same? Hardly by gagol · · Score: 1

      If you bothered to red the paper, you would know it is more of a social essay than an engineering one.

      --
      Tomorrow is another day...
  11. No because by Anonymous Coward · · Score: 5, Insightful

    Are you making kiddie porn? Laundering money? Spreading hate? Do you want the terrorists to win?

    Because I don't want every goddamn marketer out there trying to sell me their shit. I don't want to have to deal some horseshit like this because businesses feel entitled to stick their noses into my business.

    No, you are NOT offering me "convenience" - you are prying.

    As it is, I CAN create a dossier that would make an East German Stazi agent cream his pants by just hitting the credit bureaus, Google, ChoicePoint, ISPs, Cell phone companies, and every other business entity out there that has this need to collect consumer data.

    Something to hide?

    Well, just ask the atheist, gay or lesbian, peace protestor or Muslim who has their identity known what happens to them.

    The uncle of the Marathon bombers who had his face plastered all over the place is headed for some serious shit. You just know that folks are going to vandalize his house, harass him, and give him a lot of shit just because he's related to those kids and a Muslim.

    People are hateful, ignorant, cruel, shallow and just stupid - until proven otherwise. Therefore, it is imperative to keep one's secrets.

    1. Re:No because by Anonymous Coward · · Score: 0

      POTD

      I bet someone who orders high end men's shaving products, and then shops for $150 designer shirts, will soon be seeing online ads for gay wedding consultants mysteriously popping up on the news sites.

    2. Re:No because by VortexCortex · · Score: 2

      POTD

      I bet someone who orders high end men's shaving products, and then shops for $150 designer shirts, will soon be seeing online ads for gay wedding consultants mysteriously popping up on the news sites.

      Meh, I just use this VPS proxy for Slashdot and Google gives me ads for Tentacle Grape Soda and Russian Mail-Order Brides.

  12. People called cows, they have left the barn... by Anonymous Coward · · Score: 0

    How do you say "The cow's out of the barn," in Latin? There are already too many computers around. Yes, more people are using "the cloud" with mobile, networked devices, there will always be valid reasons for using standalone machines. It's a little soon to start worrying about the gubmint coming to take away our computers. Other than that, it's an interesting take on the development of computing over the years.

  13. Giving up the dream by Anonymous Coward · · Score: 3, Interesting

    There was a time when we expected computers to become so easy that everyone could use them. We've given up that dream. Now it's all "managed" again. There are admins and users again, and the admins (or their bosses) decide what the users can do and how. Computing is no longer done with a device you own but a service that someone else provides to you. Yes, you still pay for a device, but that's merely an advanced terminal.

    I blame the users. If they bothered to learn even a little about how things work, they wouldn't give up their freedom so easily. The complacency is staggering. Even people whose job depends on being able to efficiently work with computers often perform repetitive tasks manually instead of learning how to use more of the program they're working with. Of course, with users like that, who refuse to learn how to use what capabilities are already at their disposal, there's a market for the simplest automation performed as a service.

    1. Re:Giving up the dream by Anonymous Coward · · Score: 1

      There was a time when we expected computers to become so easy that everyone could use them.

      And now they have.

      What...you don't actually think that thing everyone carries in their pocket or purse is a telephone, do you?

    2. Re:Giving up the dream by ColdWetDog · · Score: 2

      I blame the users. If they bothered to learn even a little about how things work, they wouldn't give up their freedom so easily. The complacency is staggering. Even people whose job depends on being able to efficiently work with computers often perform repetitive tasks manually instead of learning how to use more of the program they're working with. Of course, with users like that, who refuse to learn how to use what capabilities are already at their disposal, there's a market for the simplest automation performed as a service.

      OK, so the Eternal Mainframe meets the Eternal Summer?

      --
      Faster! Faster! Faster would be better!
    3. Re:Giving up the dream by Anonymous Coward · · Score: 0

      No, it's a terminal which provides access to services provided by Google and Apple. Mobile phones are very much consumer devices, not general purpose computers. I know that to someone knowledgeable in the art of programming, they look and in fact are freely programmable, but that's not how they're used.

    4. Re:Giving up the dream by CAIMLAS · · Score: 1

      We've given up that's because that dream was unrealistic. It wasn't even so much a dream as it was a marketing campaign from Apple and Microsoft, and it was long before the concept of a global computer network accessible from every device was even a glimmer in the conceivers' eyes.

      You also seem to be missing the point that pretty much everyone has a smartphone and/or a computer these days, and that they use them and do things with them which were wildly impossible when that dream was commonplace. That dream was based on the concept of everyone being a creator; not everyone has that capability, most people are simply consumers.

      You seem to miss the crucial reason why things are the way they are: it's because users DO NOT want to know how things work. They want the software and devices to work and do the work they need to do for them as much as possible so they can get on with their day, on with their lives.

      Not everyone is a geek.

      --
      ~/ssh slashdot.org ssh: connect to host slashdot.org port 22: too many beers
    5. Re:Giving up the dream by Anonymous Coward · · Score: 2, Insightful

      There was a time when we expected computers to become so easy that everyone could use them. We've given up that dream. Now it's all "managed" again.

      I blame the users. If they bothered to learn even a little about how things work, they wouldn't give up their freedom so easily.

      Unbelievable. You blame the USERS for computers being hard to use.

      You just demonstrated what went wrong with computing. Right. There.

      Computer geeks: your libertarian paradise is not what other people want or need in their daily lives.

    6. Re:Giving up the dream by Anonymous Coward · · Score: 0

      Tron fights for the users!

    7. Re:Giving up the dream by Anonymous Coward · · Score: 0

      If you make something idiot proof, someone will just make a better idiot. I'm not faulting users for not understanding everything about computers. I'm faulting users for not even trying to understand the parts that would make their jobs easier. There is certainly no shortage of badly written software with bad user interfaces, but the stuff that users get stuck on is so ridiculously simple that the only reasonable explanation is that they truly reject the concept of learning how it works.

    8. Re:Giving up the dream by LoztInSpace · · Score: 1

      That's pretty much true of every computer. Just because someone only ever uses Outlook, Word, Excel & IE doesn't mean it's not a computer (or a consumer device for that matter).

  14. and the mainframe never went away by iggymanz · · Score: 2

    the most important of the world's business has always been done by mainframes, most of your money is information in a network of mainframes.

    1. Re:and the mainframe never went away by Nerdfest · · Score: 1

      It's purely because that's what was available when the systems were originally made and it's still hugely expensive to replace those systems. Many banks have and are enjoying cost savings, but they needed to bite the bullet and convert from difficult to maintain COBOL systems. Besides the cost, banks are also averse to risk, and change causes risk.

    2. Re:and the mainframe never went away by cellocgw · · Score: 4, Insightful

      Besides the cost, banks are also averse to risk, and change causes risk.

      Wait a minute: did you somehow sleep through 2008? Banks love risk, so long as it's someone else's money they're churning.

      --
      https://app.box.com/WitthoftResume Code: https://github.com/cellocgw
    3. Re:and the mainframe never went away by iggymanz · · Score: 1

      no, you have some misconceptions. there is no such need and no cost savings and to move away from mainframes. You assume a mainfrme must be running COBOL.

      Mainframes run modern software. They run it extremely cost effectively for the throughput they have, moreso than any other platform. The run it with extreme reliability and uptime. They run modern DBMS, they run enterprise java and all other modern languages, they can and do run Linux and Linux business apps. they can run x86 software on X blades.

  15. troll story is trollish by Anonymous Coward · · Score: 0

    Clearly this noobin has never done a day of sysadmin in his life on either mainframe or non-mainframe systems. Why are we listening to this guy anyways?

  16. High-functioning autists ahoy! by Anonymous Coward · · Score: 0

    Because otherwise why's all the comments up here like "Yeah, it walks and quacks like a duck, but does it have a bill like a duck? Nope! Author's an idiot. Also, why the duck are we discussing waterfowl, mainframe's a computer, you silly person!"?

    The point is, instead of making data and programs decentralized and under users' control, we're back to the Cult of Mainframe, with black boxes behind locked doors keeping and processing our data in the ways known only to the priesthood.

    The fuck's with all the "Mainframes didn't even look like this!!111" comments?

  17. Rinse, lather, repeat by bryan1945 · · Score: 4, Insightful

    Not networked, networked, not, networked, on and on. Each cycle begets a new cycle. Now it's just called "the cloud."

    --
    Vote monkeys into Congress. They are cheaper and more trustworthy.
    1. Re:Rinse, lather, repeat by iggymanz · · Score: 1

      nonsense, mainframe have been networked for decades. they can do "cloud computing". they've never gone away, they run all modern languages, dbms, and can even run Linux. now they even have expansion chassis that can take x86 blades for softwares that can't run on Z.

    2. Re:Rinse, lather, repeat by lightknight · · Score: 2

      Well, there's a reason for that. MS has this weird idea of taking the fight to other guy on their turf...no matter how expensive it is.

      Case in point: Netscape got its ass handed to it when MS took it on. Why? Because MS owned the last mile, and could afford to give away a browser for free. That hurt Netscape. Then Netscape starts loses control on the server side of things. Boom. How did MS win? By having Netscape fight on MS's turf: Windows.

      Case in point: MS wants to replace Google with Bing, Firefox / Chrome with IE. Their solution? Let's move Office and everything to the web, so Google (which lives on the web, and counts it as its home turf) has MS right where it wants it. A smarter strategy would be to leverage local machine resources to do things that can't be offloaded to Clouds / servers over the internet, and punish Google in the process.

      Seriously. MS trades a local CPU with multiple cores at multi-Ghz speeds, gobs of RAM, possibly a SSD, and more than likely a half-decent GPU for...well, a fast connection in the US (FIOS) on average might be 50 Mbps...to some tethered servers which are probably running low-power CPUs and lack GPUs...don't have SSDs...might have a lot of RAM...and more than likely, much higher latency. It's like going to Mars for a cup of water...got plenty of it here on Earth.

      Actually, given how the Tech Sector has been run over the past few years...are we being punked? We are being punked, aren't we?

      --
      I am John Hurt.
    3. Re:Rinse, lather, repeat by meta-monkey · · Score: 1

      Uh, no. We, the coders and consultants are making bank, implementing the stupid decisions passed down by management.

      "What's that boss? You wanna cloudify our web-a-spaces? And right after we ported all our offerings to 3 different mobile platforms. So you saw this cloud thing article in the inflight magazine, huh? No problem boss, just keep them pay checks coming..."

      --
      We don't have a state-run media we have a media-run state.
  18. You don't know mainframes. by Anonymous Coward · · Score: 1

    Mainframes are FAR more flexible than the x86 server farms.

    They are also less expensive for the same amount of computation - above a minimum amount. A small IBM z system can run more VMs than the equivalently expensed X86 farm, with dynamic load balancing as well.

  19. Mainframes and server farms the same? Precisely. by Anonymous Coward · · Score: 2, Insightful

    The general thinking of comparing the two is that both systems are the ones running running the show, storing the data, and being accessed by dumb-clients that only serve as terminals.

    Obviously server farms and mainframes are very different from a back-end technology standpoint, but from a viewpoint of the user they are identical in every single way. You log in with your user specific credentials, you do your work using the server's processing power and save your work in the servers storage medium. Your client likely is even set to network boot from a server supplied boot image via PXE. If your local machine is nothing but a terminal to access the backend machine, then you are for all practical purposes operating in a mainframe environment.

  20. Cloud Computing by morcego · · Score: 1

    No shit. Every time I heard someone saying he plans on building a private cloud on his computer, I ask myself why he just doesn't buy a mainframe.

    I mean, not every server farm or server room can be compared to a mainframe. But these days, when companies have VMWare clusters and what-ever clouds, it is impossible not to draw a comparison since, functionally (and sometimes structuraly) they are pretty much like mainframes.

    --
    morcego
    1. Re:Cloud Computing by iggymanz · · Score: 1

      no, not functionally like mainframes at all. they are like a network clustered bunch of x86 pc's with shared storage. that's all. Each blade is going to be bottlenecked by its few SAN links, each is ever only going to be able to give a VM at most its full CPU core count and nothing more (can't run a single vm across multiple blades for performance improvement), if a blade fails without warning then HA will take a while to spin up and continue on another blade

    2. Re:Cloud Computing by tarpitcod · · Score: 1

      Take someones description of their cloud computing service and compare it to the concept of the computing utility - like the MULTICS people talked about, it's pretty damn similar.

      Some idiots will claim it's not - that they can get to their data! Which is total garbage because while technically they might be able to get to their data, it sucks to empty a swimming pool via a straw which is what the bandwidth of an internet connection is like when there's a tonne of data in the cloud.

      So then they will say 'run your queries in the cloud', well that's awesome until the problems don't map efficiently to the topology the cloud vendor went for.

  21. You can take my pc... by Anonymous Coward · · Score: 0

    When you pry it from my fat, greasy hands.

  22. Mainframe benchmarks by Anonymous Coward · · Score: 0

    The laughable thing is, the mainframe occupying the whole room is less powerful than one blade of that server rack. The processor is so laughable IBM won't let you benchmark it against an PC, and the storage subsystem is just a bunch of stock PC disk parts.

    MAN THEY ARE SLOW!

    1. Re:Mainframe benchmarks by tarpitcod · · Score: 2

      They weren't always. Some model 360's were pretty decent. The CDC 6600 while called a 'super computer' nowadays was really a 'Large Computer'. It was a mainframe. The problem with mainframes is the same problem with every computer out there. The latency wall. There were only a few companies that really pushed the physics. That stuff has stopped at the 'system' level to a large degree. You see a few companies playing with the interconnect topology but it's not really pushing the physics stuff.

      If you take the ratio of compute to I/O of any typical modern server it's horrendously bad. To anyone out there who thinks their x86 rocks - a few simple questions:

      1) What's the ratio of memory bandwidth at various levels to I/O bandwidth? Compare that to a Mainframe from the 60's.
      2) How long on a typical server would it take to swap out all of memory? You can use SSD if you want.

      Hint) You will find 2) is many seconds to minutes for a decent sized x86 server even with SSD's. That IBM mainframe could maybe swap out all its memory in less than a second or a second or two.

    2. Re:Mainframe benchmarks by Anonymous Coward · · Score: 0

      You're an idiot. You obviously have no idea about mainframe architecture. So just shut the fuck up.

  23. Every 5 to 10 years... by Bill_the_Engineer · · Score: 5, Interesting

    Someone in the industry realizes that computing is really iterative and what's old will eventually become new again.

    I believe the origin of this periodic realizations is as follows:
    (I intentionally used "jargon" instead of "technique", since the need to create a new term doesn't seem proportional to the actual change in implementation)

    1. A college fresh out get hired at a I.T. farm armed with a new set of computing jargon that impresses human resources.
    2. He applies his version of how things should work to the current workplace and things progress well.
    3. Over the next few years the department grows and new hires are brought in to help meet demand.
    4. The new hires start preaching their version of computing jargon that was created by academia to publish a paper.
    5. The once college fresh out comes to the realization that the new computing jargon are practically synonyms for the previous generation's jargon.
    6. The new hire proceeds to step #1 and the circle of I.T. begins anew.

    The neat thing about this iterative process is that the difference in implementation of the jargon between generation N and N - 1 are small enough to not seem that much different. However the difference in implementation of jargon between the current generation and the people hired 5 to 10 cycles prior can and usually are dramatic.

    I entered the field when distributive computing and storage with localized networks were being created and evangelized. Scientific computing had to be performed at universities and anything serious had to be done by renting time on a supercomputer connected via the internet. Medium sized businesses had to rent time on mainframes to perform payroll or hired firms specializing in payroll which still exists today. Small businesses had no access to computing until personal computers and single user applications came into use. Because of the newer businesses being more familiar with distributive computing than centralized computing, they scaled personal computers up to meet the new demands. This ability to scale computing power up allows the company to grow the computing infrastructure as needed. This was not possible with mainframes. Eventually the company grows to the point that it needs to have their data and application centralized and use data centers to handle the load.

    If you step back and look solely at the physical structure (e.g. data center, clerical offices) it resembles the centralized computing from 50 years ago. However if you look at the actual data and computing flow you'll see that its a hybrid of central and distributed computing that was not imagined in the past 20 years. It's more fractal in nature. Your computing at any given moment can be centralized to your terminal, your home, your office, your department, your company, or even global (e.g. Google, Github).

    I declare this to be known as BTE's law. ;)

    --
    These comments are my own and do not necessarily reflect the views or opinions of my employer or colleagues...
    1. Re:Every 5 to 10 years... by cyber-vandal · · Score: 1

      This ability to scale computing power [...] was not possible with mainframes

      Are you having a laugh?

    2. Re:Every 5 to 10 years... by Bill_the_Engineer · · Score: 1

      In context of money required. Small businesses normally couldn't afford one mainframe much less more than one.

      --
      These comments are my own and do not necessarily reflect the views or opinions of my employer or colleagues...
    3. Re:Every 5 to 10 years... by Bill_the_Engineer · · Score: 1

      You are correct that I should be more explicit about the relationship between scalability and expense. Smaller computing platforms can be scaled up in smaller steps requiring less money up front than mainframes.

      --
      These comments are my own and do not necessarily reflect the views or opinions of my employer or colleagues...
    4. Re:Every 5 to 10 years... by Bill_the_Engineer · · Score: 1

      I may have to be more explicit by stating that the computing unit size of small computing platforms is such that allows scalability in power from the end-user up. This scalability is accessible to more people and therefore encourages a change in the field of computing.

      The unit size of large computing platforms (i.e. mainframes) do not encourage such scalability. Multiple mainframes in a data center may increase data processing power at that central location but doesn't encourage a shift of computing power downwards and looks basically the same to most of the people making use of the extra power.

      --
      These comments are my own and do not necessarily reflect the views or opinions of my employer or colleagues...
  24. Idiot by DaveAtFraud · · Score: 1

    So, what does this mean for the future of fully functional, general purpose, standalone computers? 'Offline computer use frustrates the march of progress,' says Winestock. 'If offline use becomes uncommon, then the great and the good will ask: "What are [you] hiding? Are you making kiddie porn? Laundering money? Spreading hate? Do you want the terrorists to win?"'

    Almost all of his examples are a complete non-sequitor. How does one launder money, spread hate, help the terrorists, etc. with a computer that is NOT connected? Likewise, does this mean that all of the people who owned computers in the pre-Internet era only owned them to do these things? And none of this follows from comparing mainframes and server farms or even has anything to do with mainframes and server farms.

    Sounds more like he has a guilty conscience about doing these same things from his not connected computer and should be hauled before some secret tribunal to answer for his crimes (Oh, and just denying that this is how he came to his conclusions is just further proof that he is a lying to conceal his guilt).

    How's that for a little non-sequitor inuendo?

    Cheers,
    Dave

    --
    They that can give up essential liberty to obtain a little temporary safety deserve neither safety nor liberty.
    Ben
    1. Re:Idiot by Anonymous Coward · · Score: 2, Informative

      Have you read TFA? He's not advocating going back to the mainframe-terminal paradigma. He's warning against what is commonly perceived as the way things will inevitably go. If you're seen as "out of the norm", you're ostracized. This is reality. You don't do what everything does - and is recognized as "good" because that's what everybody does - you're "stranger danger". And in this day and age, being "stranger danger" may be a death sentence.

    2. Re:Idiot by Anonymous Coward · · Score: 0

      Although there might be a hint of hyperbole in that, I can at least agree that being different in any way shape or form could harm your career if you're not careful.

      At my job, there are a lot of religious conservatives. I've been here long enough to identify who I can trust, and so far I have only told one person outright that I am atheist, and I have told no-one that I think Obama might be an OK president. My boss is a dyed in the wool "Reaganite" with calendar of his favorite president hanging on the wall, so I know what not to share and who to keep it from.

      Thankfully, they have not asked to be Facebook friends with me, and the one person I offered it to declined out of caution (which I totally understand).

      If CISPA is passed, I wonder how it'll be abused. Under the current leadership I'm not afraid of them knowing what my religious affiliation is, or what kink I have. But, in the future if we fall under the grip of religious indoctrination and an administration with purely fascist tendencies, I think I would be very afraid. Among my greatest fears has to do with my old political internet posts here on Slashdot and other places where they cannot be deleted. If we ever go full Nazi, the government will have no trouble finding people and throwing them in gulags for past transgressions in statements from times long since past.

      Also, did you know that Windows 8/RT has a special hidden cloud just for apps? Its intended to help apps restore from their previous state in the event of a crash, as well as to share that state with an individual's various computers and tablets.

      If you're a perv who looks at something deemed illegal by the state, those images could possibly be run through a special filter on-the-fly, creating a highly efficient way to nab perverts, throw them in jail, and force them to live with a scarlet letter for the rest of their days.

      No one supports viewers of child-porn. There's no sympathy for them. But who can say that the dragnet couldn't be expanded to cover other kinds of pornographic material? Maybe the greatest threat to God Fearing America will be furry porn? Maybe it'll be the fat fetish? maybe it'll be your own little secret indulgence. It harms no-one, you'd prefer to keep secret, but the government may someday disagree... and if they do you could be screwed.

      You are safe for now. In this "day and age" - barring complicated circumstances like a murder investigation - you don't really have to fear death sentences from your computing activities, but the internet is pretty good at remembering, and politics change with the times. What's safe today could be a death sentence for you tomorrow. It doesn't even have to be pictures. It could be text like the above.

  25. We need both client and server by WOOFYGOOFY · · Score: 1

    HTML5 interfaces will always suck just the way javascript HTML 4 interfaces suck- you can't take a server hit every time you want to react to a mouse movement or process a keystroke.

    For a large number of apps, this actually doesn't matter but for people who really do creative work with their computer , the UI and a very large amount of processing of local data will have to take place on the local machine.

    I suppose their are entities out there actively plotting the end of personal general purpose PC but to say that they somehow control what direction the world will go is paranoiac to the well-known Kaczinsky Limit.

    Benefit, productivity, competitive advantage, goodness, fun whatever gets maximized when the cloud/server/web / whatever is utilized for what its good for - communication and distribution of content and the processing of truly HUGE data sets or data from a very large number of data sources. .

    On that last point, even there the necessity of the cloud is challenged by distributed applications of the SETI type.

    I think Adam Smith pretty well had this down with the idea of relative competitive advantage. Servers should not try to do UIs or make believable promises WRT to the security needed for very critical data. Local PCs are easily connected to a wide range of other-generated / other processed data via the web and servers.

  26. Rubbish, about 1/5rd an i7 performance by Anonymous Coward · · Score: 1

    Top of the range IBM kit, the multi-million dollar kit is 17802 MIPS, about 1/5th the performance of an i7 based server.
    http://www.computermerchants.com.au/zseries-mips-chart-update-july-2008/#.UXP6LcpdkcE

    IBM sales men will tell you all manner of lies, but they won't let you benchmark their mainframes, against PCs and they don't for a damn good reason.

    The disk subsystem is assembled from PC parts (IBM sold their disk division), the processor is a slow IBM model (IBM haven't been a threat to Intel for ages), and when pressed on performance, the salemen usually pretend the MIPS are more powerful ops, more secure ops, as if multiplying integers is somehow special on mainframes.

    It's pitiful.

    You talk into your phone and SIRI does a voice analysis, and searches larges data sets for an answer. Watson had to be fed the question as text on Jeopardy, because there was enough processing power to do voice to text.

    1. Re:Rubbish, about 1/5rd an i7 performance by Nerdfest · · Score: 1

      ... and you pay for using those processors in addition to the cost of the machine.

    2. Re:Rubbish, about 1/5rd an i7 performance by Anonymous Coward · · Score: 0

      You're an idiot.

    3. Re:Rubbish, about 1/5rd an i7 performance by Jeremy+Erwin · · Score: 1

      Watson was made up of 90 Power 750 servers (2880 cores, 16 TB memory, etc). Benchmarks for this sort of system are readily available The iPhone offloads its voice recognition to an Apple "mainframe".

    4. Re:Rubbish, about 1/5rd an i7 performance by Anonymous Coward · · Score: 1

      > as if multiplying integers is somehow special on mainframes

      Uh, because it is?
      The multiplication is done modulus 3 in parallel, and the result is then compared, if it fails it is retried and/or moved to another processing unit and the broken one is disabled.
      Does this effort make sense? Most of the time not. But is it special? Yes.

    5. Re:Rubbish, about 1/5rd an i7 performance by Anonymous Coward · · Score: 0

      you are an idiot.
      I read through your analysis trying to figure out how your comparison worked. while parts are technically true, such as the i7 being a more powerful chip, power is more than a cpu arch...

      but that is not why I am calling you an idiot. I am calling you an idiot (and charitably, at that) over the last remark. first, your phone is not running siri. a big server farm runs siri. second, the phone is not an i7 so the comparison is basically like comparing you to someone informed. third, Watson was not a demo of speech to text functionality. it was fed the text version to avoid ambiguity. ask siri "which homes did holmes live in?" and you will get back a response nearly as cogent as your post. that is to say, "rubbish". now, please, go learn something then come back and talk.

      again: you are an idiot.

    6. Re:Rubbish, about 1/5rd an i7 performance by Jeremy+Erwin · · Score: 1

      From some IBM propaganda

      The z196 is the premier high end server and the flagship of the IBM systems portfolio. It contains 96 of the world’s fastest, most powerful microprocessors running at 5.2 GHz and is capable of executing more than 50 billion instructions per second. With up to 80 configurable processors, the z196 can scale to over 52,000 MIPS (Millions of Instructions Per Second) of compute capacity in a single footprint.

      Hurray for multiprocessing!

  27. How did they know?! by Anonymous Coward · · Score: 0

    If offline use becomes uncommon, then the great and the good will ask: "What are [you] hiding? Are you making kiddie porn? Laundering money? Spreading hate? Do you want the terrorists to win?"

    Yes, all of those. Brainwashed scantly dressed girl scout sleeper cells. Their leader, the cookie master, commands them to rain door-to-door jihad on the snackers.

    Now that I've exposed my knowledge of them, it's only a matter of time until I'm cooked. My only hope is that the New Boss will reach me first and be merciful!

    1. Re:How did they know?! by chromas · · Score: 1

      And you did all that while your computer was offline, right?

  28. Rudolf Winestock for President by LifesABeach · · Score: 1

    'Offline computer use frustrates the march of progress,' says Winestock. 'If offline use becomes uncommon, then the great and the good will ask: "What are [you] hiding? Are you making kiddie porn? Laundering money? Spreading hate? Do you want the terrorists to win?"'"

    Really? I think the Tea Party has found their next candidate for president. Now if only he had a personal life like the "Newt."

    1. Re:Rudolf Winestock for President by Voline · · Score: 1

      Slow down. Winestock is not making the "If you're offline you must have something to hide ..." argument, he's anticipating it. He's warning that this is an argument authoritarians will soon be making and so one should be ready to defend the right to even have a general-purpose computer and keep one's data locally.

    2. Re:Rudolf Winestock for President by LifesABeach · · Score: 1

      Ups, I missed that one.

  29. IT needs some kind of an apprenticeship system by Joe_Dragon · · Score: 2

    IT needs some kind of an apprenticeship system or at lest more tech schools where you learn from people who have done real work and not so much people working on there academia papers and you have more hands on learning as well.

    1. Re:IT needs some kind of an apprenticeship system by Bill_the_Engineer · · Score: 1

      I meant this progression to be a "good thing". One problem with apprenticeships is that you reinforce the established way of doing things. Bringing people in from the outside, especially those who learned from others that read or write academic papers allows new concepts to be integrated with established practice.

      --
      These comments are my own and do not necessarily reflect the views or opinions of my employer or colleagues...
    2. Re:IT needs some kind of an apprenticeship system by Anonymous Coward · · Score: 0

      These are called "trade schools"...

      And there is a reason these graduates produce some of the most insecure and sloppy code ever seen.

  30. Valid Observation, Poor Argument by Capt.Albatross · · Score: 3, Insightful

    Mr. Winestock's parallels between server farms and mainframes are reasonable, if unoriginal, and the same can be said for his concerns over privacy and social control. His attempt to claim the former as the causative agent for the latter, however, goes wrong right from the start: 'Mini/micro-computers were supposed to kill the mainframe.'

    Not so. They came about firstly because technological advances made them possible, and also because some smart people realized that they would allow us to do things that, in practice, we could not do before. The pioneers of these developments were not interested in reproducing, much less replacing, mainframe computing.

    Turing showed us that the form of our hardware doesn't dictate what we can do with it. To understand the arc of privacy erosion and social control, we need to examine social history and human nature, not the artifacts of technological advance.

  31. The terrorists have long ago won by Anonymous Coward · · Score: 0

    So to win I too must become a terrorist.

    Now for some snazzy mirrorshades. I already have the mullet.

  32. Download caps / lag / 3g, 4g, LTE roaming costs wi by Joe_Dragon · · Score: 1

    Download caps / lag / 3g, 4g, LTE roaming costs will make it very hard to go all back end with your system being just a dumb terminal.

    And with roaming cost that can hit $10-$20+ a meg in Canada (higher in other places). A nice remote desktop at least 1024X786 can burn data fast.

  33. The only way to boost popular techno-literacy by Anonymous Coward · · Score: 0

    Somehow it needs to act as an aid to getting laid instead of a barrier. If we can figure out how to do that the problem will fix itself.

    1. Re:The only way to boost popular techno-literacy by Anonymous Coward · · Score: 0

      Eliminate all females, and replace them with androids? Perhaps we could understand them, then :)

    2. Re:The only way to boost popular techno-literacy by Anonymous Coward · · Score: 0

      Except that those would be gyndroids. An- is male.

  34. GOOGLE DOESN'T CARE ABOUT ERRORS !!! by Anonymous Coward · · Score: 4, Insightful

    "Google, for example, has pretty full-featured job control layered on top of their server farm."

    Google has never cared about errors.

    Who gives a damn if what absolutely positively SHOULD have been the very first result is instead the fourth or the fifth result, or if it appears on page two of the results, or if it somehow magically disappears into the ether because commodity server #XJ42 in rack #43HB on aisle #521JJ in column #447F in building #QQZ1 in server farm #H61M happened to have crashed just as the query response was being assembled?

    Especially if the query involved "Justin Bieber", "Lindsay Lohan", or "Natalie Portman Hot Grits".

    IBM, on the other hand, has always cared about errors - has always, in fact, been FANATICAL about errors.

    If you send a query to an IBM mainframe, then you're expecting umpteen-sigmas of confidence that the mainframe will actually be up and running, that you'll get an actual response, and that the response, when it finally arrives, will be 100% CORRECT.

    Especially when the response is something along the lines of "DANGER: CHILD KNOWN TO BE ALLERGIC TO AMOXICILLIN. ALLERGIC RESPONSE INCLUDES ANAPHYLACTIC SHOCK. PRESCRIPTION REQUEST THEREFORE INVALID AND REFUSED."

    1. Re:GOOGLE DOESN'T CARE ABOUT ERRORS !!! by Runaway1956 · · Score: 1

      As far as I know - Google isn't running prescription services on their search engine servers. Do you know something that I don't?

      Are you suggesting that all medical services are run on IBM servers? I find that a bit hard to believe. Those hospitals and doctor's offices that I've been in seem to run Windows on Dell machines, almost exclusively. I've seen a couple of computers with something that looked like it might be Solaris, but I didn't actually have access to the machines, and couldn't investigate.

      I suspect that the computers used to run various scanning equipment might run versions of Linux, but I have nothing to base that suspicion on - other than the fact that all of the embedded equipment at work is based on Linux.

      Yeah - I'm perfectly aware that what I've seen is NOT THE SERVERS - but the fact that everyone seems to opt for Windows-centric Dell equipment at the front offices might indicate that they run the same stuff out back.

      --
      "Windows is like the faint smell of piss in a subway: it's there, and there's nothing you can do about it." - Charlie Br
    2. Re: GOOGLE DOESN'T CARE ABOUT ERRORS !!! by santiago · · Score: 3, Insightful

      You have no idea what you're talking about. Dropping the occasional search result is fine, but what about failing to record billing for the ad system, dropping mails you were supposed to receive in your Gmail account, or failing to save the doc you were editing? Google does a lot more than serve search results, and most of that needs to work every single time.

      The fact of the matter is that even the most expensive hardware eventually fails, so your software needs to be able to deal with it and fall back to working units. Once you've written your software to handle hardware failures, you can run on really cheap hardware. And, it turns out that buying a lot of really cheap computers some of which are broken all the time gets you way more computing power than trying to buy a few really robust machines.

    3. Re:GOOGLE DOESN'T CARE ABOUT ERRORS !!! by Anonymous Coward · · Score: 2, Interesting

      "Google, for example, has pretty full-featured job control layered on top of their server farm."

      Google has never cared about errors.

      Actually, Google cares about errors so much that they have invented some nice error correction techniques. For example, in Hadoop and the original Google's equivalent system, computers are allowed to just break in the middle of calculations. All that is transparent to the high level code that runs on the system; the program gets executed normally, perhaps with some delays but otherwise uninterrupted at all.

    4. Re:GOOGLE DOESN'T CARE ABOUT ERRORS !!! by thejynxed · · Score: 1

      They don't. Also, what should scare you, is the scanners that run customized versions of Windows XP Service Pack 2, and are directly accessible on the internet via the external to internal network bridge.

      Yes, I have seen this at hospitals. As for their medical records departments, etc. That's all mainframe stuff, and quite a bit of it is even kept on tape.

      --
      @Mindless Drivel: 100% of Twitter posts ever Tweeted.
    5. Re:GOOGLE DOESN'T CARE ABOUT ERRORS !!! by Anonymous Coward · · Score: 0
      You can run terminal emulators on Windows. You can be almost guaranteed that most hospitals patient administration systems still run on some type of mainframe.

      Downstream systems (the systems that you see doctors and nurses use) these days are most likely not on a mainframe but have an interface to the patient administration system that does the patient bookings, bed movements, discharges, allergies, demographic information, etc. when the patient administration system gets updated with new patient data it'll send messages (HL7) over the interface to the downstream systems that run on your typical windows machines and additional functionality specific for the department is provided by these systems. You can have hundreds or thousands of the downstream systems but normally only one patient administration system and it's normally on a mainframe.

    6. Re:GOOGLE DOESN'T CARE ABOUT ERRORS !!! by Anonymous Coward · · Score: 0

      Yet you are perfectly aware of what you have seen, I happen to be typing on a Dell computer at my job desk, running firefox over win7 and, whoa, IBM Personal Communications (http://www-03.ibm.com/software/products/us/en/pcomm)...

      What a surprise, huh?

    7. Re:GOOGLE DOESN'T CARE ABOUT ERRORS !!! by Anonymous Coward · · Score: 0

      With all the improving reputation regarding on the web reward purchasing, it's got grow to be extremely less difficult for many to go properly with out losing their particular work-time. Furthermore these kinds of internet vendors not merely give you the particular simplicity of purchasing yet additionally, it gives http://www.ornaments-giftshop.com/ you a fantastic probability to examine the costs recommended with diverse models of aircraft web sites. In addition, these kinds of on the web reward retailers furthermore offer you special discounts during specific situations just like Romantic days celebration, Dad's evening, Single parent's evening and also about different fairs furthermore. Thus, this kind of on the web reward purchasing could possibly be less costly alternative when compared with in which regarding terrain retailers.
      Purchasing is significantly diverse nowadays as compared to inside earlier several years. As a result of on the web reward retailers you will find fantastic reward alternatives with a single spot. Nowadays you'll find so many on the web reward retailers accessible together with massive selection of reward alternatives. Special birthday is probably the specific situations for many, thus these kinds crafts of internet vendors retain a particular set up or even a plane model part regarding special birthday items. In addition to normal items just like cologne, diamond jewelry, observe or perhaps accent merchandise, in addition they send out special birthday cakes and also biscuits to be able to which you would like to current. For your capability of an individual, the web reward retailers grouped their particular gifts beneath diverse classes; for instance items regarding your pet, the girl, youngsters, father and mother and so forth. Every one of these ensure it is more hours conserving and also helpful.

  35. But CS is not IT and people who do academic papers by Joe_Dragon · · Score: 1

    But CS is not IT and people who do academic papers are type of people in IT who have been in academic for most of there life and have little to no hands on IT work.

    And we don't need more people in IT loaded with academic smarts but little IT book smarts / Little hands on smarts.

  36. And not just hardware by Anonymous Coward · · Score: 0

    If anyone has been paying attention, its quite clear that lots of new "tech" are just reiteration of old 70's & 80's unix functionality in the "core" level ..

  37. Painfully obvious by Anonymous Coward · · Score: 0

    This guys knows nothing about actual mainframes. His long, rambling post may be interesting for other reasons.

  38. Re:But CS is not IT and people who do academic pap by Bill_the_Engineer · · Score: 1

    I'm not so sure that is entirely true. I do not believe data centers (at least the good ones) are complete void of CS people. IT people who maintain an infrastructure do not work inside a vacuum. They are either influenced directly by CS people within their organization, or indirectly by the computing appliances or applications that they maintain.

    I'm afraid a lot of low level techs become scape goats for high level techs that should have known better. I thought we already have tech schools that train low level IT workers. The ones at my place of work have trained at these institutions and are continuously trained through vendor provided seminars and in-house training. I don't think the shortage is entirely from the lack of talent or pre-occupational training. I think a lot of problems can be attributed by the lack of continuing training while employed. I know of several companies which hired techs to do nothing but keep the network up. They don't provide training and their requirement to run a legacy network running a custom built application several years old are satisfied. However when compared to their colleagues that work elsewhere they begin to look like IT lay people who are ill-equipped to handle modern applications or security requirements.

    Most these businesses keep their in-house IT far way from the internet except for the email that is provided by an off-site service. Then one day a middle manager will go to a retreat, learn how they can improve their business by integrating with social media and whatever other buzzwords are offered, returns and tells their "sheltered" IT department to make it so. This is where the news making blunders are made...

    --
    These comments are my own and do not necessarily reflect the views or opinions of my employer or colleagues...
  39. serving a recall notice on "Don't Be Evil" by epine · · Score: 1

    "If you have something that you don't want anyone to know, maybe you shouldn't be doing it in the first place."

    That statement is nothing shy of a Full Monty disgrace to free enterprise. Nobody ever has a nice thing to say about government, and this leads to the comforting illusion that we can devolve the beast of government (for the most part) to the free market where much of government's function would be better served, until some high and mighty idiot in the private sector comes around saying something like this and bursting everyone's happy bubble. Well done, Eric, running the graduated approach to managing one's personal boundaries straight to the tip heap, for the betterment of all society. Yes, this is exactly what government by quarterly report will look like when that fine day finally comes. Book it.

    There has never in history been a society that has strayed so far into the glass fishbowl: in a closed community where no behaviour goes unnoticed, living quiet lives of desperation is the order of business. Woe to anyone who dares to shirk this shackle (a theme of the very difficult movie Breaking the Waves). And yet, this too is not enough?

    What a pompous ass to make such a remark. So close, and yet so far. Google could have been so much worse. For a long stretch, their sane and (relatively) moral decisions far outweighed their missteps. Then they caught wind of Facebook eating their lunch, and now they seem hell-bent on making up for lost time. I can barely express my disgust at the implications of that remark.

    There's that old joke about Gates declared darkness "the new standard". Now we have Schmidt declaring the naked light bulb in the holding cells of the Lubyanka as the new, unceasing dawn.

    I was reading about circadian phase entrainment the other day. In the hamster model (which I say generically, forgetting the precise rodent flavour) they use constant dim light to establish the free running state (which is actually the free running state in constant dim light). They don't use constant bright light, because constant bright light causes the cells of the suprachiasmatic nucleus to lose synchrony (effectively destroying the body's internal circadian signal altogether). In the torture setting--if that is in fact the purpose of the unblinking naked light bulb hanging above arm's reach in every cell--loss of circadian rhythm would have an effect on sleep that would promptly dissolve and disintegrate all sense of perspective and self-hood. This is, of course, what they wish to achieve. One doesn't torture the whole man, one tortures the wretched shell, so that the whole man shall never take up residence ever again.

    Praise be to Google, keeper of the constant light.

  40. I think he's worried about cloud computing... by rsilvergun · · Score: 2

    being abused by gov't. I don't think it really matters. Online is still just online, and I've said before and will say again that the Occupy Wall Street Movement showed that in the real world when the gov't wants something to go away it does.

    Basically we don't really have the freedom he's saying we'll lose. Real freedom is economic freedom. You're not free as long as somebody controls your access to food, shelter and health care. Until then you'll do exactly what they say and so will everybody else.

    If you want freedom stop bothering with all these surveillance scares and start asking what it takes to really be free. Ask yourself if you can ever be free in a world where 6 people have more money than 100 million others combined?

    --
    Hi! I make Firefox Plug-ins. Check 'em out @ https://addons.mozilla.org/en-US/firefox/addon/youtube-mp3-podcaster/
  41. Is this burning... by Anonymous Coward · · Score: 0

    ...The Eternal 'Frame?

  42. Re:We need both client and server by Lennie · · Score: 1

    You know what is interresting, have you seen the new X1 from Comcast ? It really is very close to a dump terminal:

    http://www.openstack.org/summit/portland-2013/session-videos/presentation/keynote-openstack-as-a-platform-ecosystem
    http://www.openstack.org/summit/san-diego-2012/openstack-summit-sessions/presentation/open-source-versions-of-amazon-s-sns-and-sqs

    It send keystrokes one way and receives screens back.

    --
    New things are always on the horizon
  43. But we also need people who are not codeers doing by Joe_Dragon · · Score: 1

    But we also need people who are not codeers doing other IT work. Also CS can trun out people loaded with theory and other skills that are really not needed do you really need to know the full theory of file systems / and other low level OS stuff? or more of a higher level view of stuff? Ok the low level stuff is good if you are coding a OS but not so much if you are coding a app we don't want apps with there own file saving / loading systems over the build in os ones?

  44. Automation of repetitive tasks on a smartphone by tepples · · Score: 1

    Even people whose job depends on being able to efficiently work with computers often perform repetitive tasks manually instead of learning how to use more of the program they're working with.

    What...you don't actually think that thing everyone carries in their pocket or purse is a telephone, do you?

    A smartphone is a computer, but from what I've seen, its automation framework isn't quite as rich as that of a PC.

    1. Re:Automation of repetitive tasks on a smartphone by pipedwho · · Score: 1

      A smartphone is a computer, but from what I've seen, its automation framework isn't quite as rich as that of a PC.

      That's because it's pretty much already automated everything that the average user cares about. Which is basically the definition of ease of use.

      We're not quite at the stage where a general user can simply say: "computer, consider the following..." And get a functionally useful response.

      But, when that does finally happen, it is more than likely that the resulting system will not resemble today's 'desktop' computers.

  45. Last TCO I was involved with... by Dcnjoe60 · · Score: 3, Insightful

    You haven't tried the IBM kool-aid yet. Those people whose jobs currently rely on mainframe expertise are very happy with them. They do have better error-checking but everything else is at least an order of magnitude out of whack with commodity hardware price/performance, and in many cases, several orders. You can reduce some of the costs on their zSeries by buying specialised processors for DB2, Java, and Linux (~100K a pop) so you don't have to may for MIPS usage but the costs are still astronomical for the performance. If it was cost effective, don't you think Amazon would be running its cloud services on them?

    The last TCO I was involved with actually showed that the mainframe was the more cost effective approach for the use case at hand.

    As for Amazon, well that is hard to say. If when they first started, they knew how successful they were going to grow and how quickly, maybe they would have gone with a mainframe solution.That's the nice thing of TCO analysis, it eliminates, or should eminiate, any platform bias the decision makers have. Then again, it also depends on really knowing what future growth patterns and expected use cases are or it is just more GIGO.

    1. Re:Last TCO I was involved with... by bored · · Score: 1

      The last TCO I was involved with actually showed that the mainframe was the more cost effective approach for the use case at hand.

      TCO estimates are the finest snake oil money can buy. I see them periodically, and often times it doesn't take me long to change some tiny assumption (like swapping a DS8k for a lower function faster disk subsystem on the PC cause it doesn't need FICON for example) to throw the results off. Sometimes simply changing vendors is enough. The ones I'm particularly fond of can be found on IBM's site where they compare 10 year old PC clusters with the latest piece of IBM gear. Well duh, newer hardware had better be cheaper than the old stuff on a price/performance curve otherwise why upgrade?

    2. Re:Last TCO I was involved with... by Dcnjoe60 · · Score: 2

      The last TCO I was involved with actually showed that the mainframe was the more cost effective approach for the use case at hand.

      TCO estimates are the finest snake oil money can buy. I see them periodically, and often times it doesn't take me long to change some tiny assumption (like swapping a DS8k for a lower function faster disk subsystem on the PC cause it doesn't need FICON for example) to throw the results off. Sometimes simply changing vendors is enough. The ones I'm particularly fond of can be found on IBM's site where they compare 10 year old PC clusters with the latest piece of IBM gear. Well duh, newer hardware had better be cheaper than the old stuff on a price/performance curve otherwise why upgrade?

      A TCO prepared by a vendor is simply more marketing material. The TCO analysis that I am referring to is one that is prepared inhouse, using inhouse assumptions. Any organization that is large enough to be contemplating server farms or mainframes is in a position to be doing its own analysis.

    3. Re:Last TCO I was involved with... by Anonymous Coward · · Score: 0

      Did that TCO include the cost of hiring and maintaining mainframe experts? There's a premium for specialist knowledge that doesn't exist in the Linux/x86 world.

    4. Re:Last TCO I was involved with... by snadrus · · Score: 1

      Mainframes failure is in handling the volatility of today's technology world. With nothing resell-able, cheaper costs are meaningless versus piecemeal purchasing of ever-better hardware.

      --
      Science & open-source build trust from peer review. Learn systems you can trust.
  46. The blades will get smaller and smaller by istartedi · · Score: 1

    The blades in the racks will get smaller and smaller, and cooler and cooler until... a bunch of hippie hackers decide to build a server farm on their kitchen table. The "dumb terminal" will sit right there on the kitchen table too. It'll be the same "cloud" architecture, but small and private. The PC (Personal Cloud) revolution will begin.

    --
    For all intensive purposes, "whom" is no longer a word. That begs the question, "who cares"?
  47. OnLive by tepples · · Score: 1

    You know what is interresting, have you seen the new X1 from Comcast ? It really is very close to a dump terminal:

    (Two long YouTube videos)

    Are transcripts publicly available?

    It send keystrokes one way and receives screens back.

    The last time I heard about that, it was called "OnLive".

  48. Turing completeness and code signing by tepples · · Score: 1

    Turing showed us that the form of our hardware doesn't dictate what we can do with it.

    Which is why certain makers of consumer electronics are bending over backwards to use cryptography to keep hardware that they sell from being used as universal Turing machines. Apple, game console makers, and pay-TV converter box makers are especially guilty.

  49. Re:Download caps / lag / 3g, 4g, LTE roaming costs by tepples · · Score: 1

    Download caps / lag / 3g, 4g, LTE roaming costs

    Then instead of 3G and LTE, use Wi-Fi with a wired upstream. Most restaurant chains provide it to customers. Or are you talking about using a laptop on the bus?

  50. Re:We need both client and server by Kittenman · · Score: 1

    You know what is interesting, have you seen the new X1 from Comcast ? It really is very close to a dump terminal:

    Dump terminal?

    --
    "The greatest lesson in life is to know that even fools are right sometimes" - Winston Churchill
  51. Re:We need both client and server by gagol · · Score: 1

    Dump terminal: 1. A terminal that connects exclusively to /dev/null. 2. A thin client used in garbage disposal business.

    --
    Tomorrow is another day...
  52. Something is missing here... by mendax · · Score: 1

    The article talked about the "priesthood" of the mainframes. This priesthood did not only extend to the people who worked in the air conditioned room with the raised floor, it also extended to the people who GRANTED you access to the computer itself. Even on the "ancient" mainframes of the 1960's, you still had to have an account created for you. What we use today to enforce security was then used to keep track of computer (CPU, disk and tape storage, and paper usage) so the "archbishop" who ran the computer room knew who to bill. Without that account you were doomed to only look at the computer through the glass windows and drool. What the introduction of the PC did is make it possible for anyone who wanted to hack^H^H^H^Hprogram and had enough money to buy it to do so. Today, thirty-five years later, used PCs are a dime a dozen. All one needs is to buy, borrow, or steal an Internet connection.

    I frankly don't care what others want to use computers for; I want to program. It's what I love to do, it's what I'm best at, and I suspect that this is a sentiment shared by many of not most of the geeks here. I don't see the rebirth of the mainframe in the form of cloud computing to be any threat to that so long as the free software movement continues to exist and continues to produce high-quality compilers. In fact, if anything, the fact that the software engineering world is so highly specialized provides so many avenues for a true hacker^H^H^H^H^H^Hprogrammer to explore, from the very broad wastelands of the web app world all the way down to the machine level with embedded systems. In fact, any geek who wants to explore the mainframe world can do so. The Hercules emulator for IBM 360 and its derivatives (including the current Z series mainframes) is freeware and IBM wrote the operating systems and compilers for them before software could be copyrighted so those are in the public domain. So, this is a wonderful time to be a geek!

    --
    It's really quite a simple choice: Life, Death, or Los Angeles.
  53. Re:We need both client and server by Lennie · · Score: 1

    I obviously meant what you call a thin client.

    --
    New things are always on the horizon
  54. ad hominem attack much? by girlinatrainingbra · · Score: 0

    re: cheaper than paying 20 creeps with greasy hair to change hard drives, stack servers into a rack and fuck up the rollout of new VMs. [emphasis mine, pointing out the ad hominem attack]
    .
    Way to do an ad hominem attack on Linux on non-mainframe, dude! Those "creeps with greasy hair" are the Lintel guys about whom you're complaining and whom you've disparaged and thus made it harder for them to see or consider or accept your points (valid or not) about mainframes and system Z. Those guys are probably also a decent percentage of the audience here on slashdot.
    .
    If, on the other hand, you're a marketing shill and a marketing droid for IBM, then you're probably trying to reach the V.P. level decision makers and C.I.O.s, and in that case, calling those lower-level workers "creeps with greasy hair" ( as opposed to those wonderful starched-white-shirt-IBM-consultants-at-$1kph [one kilobucks per hour] who must use some ah-mazing hair-care products available exclusively from the salons of International Blowdry Machines ) makes sense. You're trying to make them feel shame for not using IBM. Way to use linux for personal gain and give nothing back to the community!

  55. Racks and cables? by the_arrow · · Score: 1

    Well that's only the outside. Once you log in to them the look like this.

    --
    / The Arrow
    "How lovely you are. So lovely in my straightjacket..." - Nny
  56. Uptime: 619778000 seconds and counting by Anonymous Coward · · Score: 0

    It's been up since September 1, of this year. Today is September 7174, 1993.

  57. I was referring to scripting by tepples · · Score: 1

    That's because it's pretty much already automated everything that the average user cares about.

    Not all users are average. Mobile operating systems still tend to lack the sort of user scripting that not-average users care about and that average users can still learn to care about.

  58. Yup. by Anonymous Coward · · Score: 0

    There's no difference between the "cloud" and a mainframe... except either it's your mainframe, or you're timesharing. Then there's the point that in a lot of the cloud, you get to allow someone else to make money off of your data.

    Around 2000, an IBM engineer maxed out a good-sized mainframe... with 48,000 instances of Linux running under VM (that's IBM VM, runs on mainframes only, has been around since the seventies). The system *comfortably* ran 32,000 instances.

    So, how many servers do you want to pay for? How many VMs will *comfortably* run on each?

    Every ten years or so, I hear that the mainframe is dead (again, like Kenny). Usually, within a few months, I also see that IBM has declared record shipments of new mainframes.

    Too many of you think we're in a "post-privacy" world; tell that to any company or government, or most people, and they'll look at you like you're an idiot.

    And you are.

    Not too many months ago, the UK decided they would *not* put things on the Cloud, because the provider couldn't guarantee that the British data would be stored, *all* of it, on UK soil. You want your nuclear data stored in any datacenter, maybe, say, in Africa, or eastern Europe, or India, or China? I work for a US federal contractor (non-military, thank you), and I had to get a "position of trust" clearance, because I'm a sysadmin. You want to GUARANTEE to me that every single person who has physical or admin access to the servers and disk farms that our data might go on HAS SOME LEVEL OF CLEARANCE? You want your personal infomation, including, say, medical and financial, secure? Oh, you don't *care* who knows enough to steal your identity, or keep you from getting a job due to medical reasons that might increase their benefit costs?

                      mark

    1. Re:Yup. by Ash-Fox · · Score: 1

      Not too many months ago, the UK decided they would *not* put things on the Cloud

      I don't believe you.

      --
      Change is certain; progress is not obligatory.
  59. PC vs Mainframe - The biggest difference is ... by Anonymous Coward · · Score: 0

    The mentality of the operators. Server admins do things in the name of expediency that mainframe operators would never do. The PC hardware is not the source of most outages. It is human error. Humans operate both servers and mainframes so what accounts for the difference? Attitude of the operators. To get things done quickly you need to think like a server admin but to keep things up and running you need to think like a mainframe operator.

  60. Re:We need both client and server by gagol · · Score: 1

    Thin client: A customer weighting less than 100 pounds.

    --
    Tomorrow is another day...