Slashdot Mirror


Microsoft Competes In Supercomputer Market

HoboMaster writes "Microsoft is releasing a public beta of Windows Compute Cluster Server 2003 in their first attempt to compete in the supercomputer OS market. Gates is planned to speak at the 2005 Supercomputer Conference, which will be Microsoft's first appearance at the conference. Gates, as always, has high hopes for this new version of Windows, even claiming it to be as powerful and easier to use than Linux."

464 comments

  1. oh boy by jst4fun · · Score: 5, Funny

    so its super blue screen(s) of death

    --
    Normal is Boring!! http://www.dealwithdeals.com/
    1. Re:oh boy by sundancekid503 · · Score: 0, Offtopic

      As soon as I saw this story I was groaning in anticipation of the flood cliche /. BSOD jokes that were sure to follow.

      Didn't take long at all....

    2. Re:oh boy by Misch · · Score: 2, Interesting

      Windows Zombie Network. I think Microsoft has already scueeded in creating large "supercomputers", of course they're oncontrolled and spit out viruses and spam... oh, where did you want to go today?

      --

      --You will rephrase your request for me to go to hell. Goto statements are not acceptable programming constructs
    3. Re:oh boy by DurendalMac · · Score: 5, Funny

      They should have just called it Windows Clusterfuck Edition.

    4. Re:oh boy by Anonymous Coward · · Score: 0

      No...That's just the "Blue" light special.

    5. Re:oh boy by sdnoob · · Score: 3, Funny

      brings to mind the age-old joke ...

      q. "why do i need to buy a new computer?"

      a. "so you can reboot windows faster."

    6. Re:oh boy by TheSpoom · · Score: 1

      Oh, they're controlled. Mostly by opped users in hidden IRC channels.

      --
      It's better to vote for what you want and not get it than to vote for what you don't want and get it.
      - E. Debs
    7. Re:oh boy by Reaperducer · · Score: 2, Funny

      A Windows product with "Cluster" in the title.

      Finally! Some truth in advertising!

      --
      -- I'm old enough to have lived through six different meanings of the word "hacker."
    8. Re:oh boy by Anonymous Coward · · Score: 1, Insightful

      A. Computers are becoming cheaper
      B. Computers are generally replaced every ~4 years
      C. People don't like throwing away something that they spent several hundred dollars on

      What happens to the old computers that someone does't want to use? If Microsoft creates a any-idiot's solution for pluging those old boxes together to make a file server, web proxy, webpage hosting box... If there isn't a market for it now, there will be by the time this cluster OS is finished.

      P.S. On the liscense comments: When Microsoft starts loosing customers because of their per processor liscense then it will be changed. They may not make the most solid software, but they know how to make money.

    9. Re:oh boy by Anonymous Coward · · Score: 0

      You don't need a supercomputer to accomplish any of those tasks!
      Why not do it with plain old Windows/Linux/FreeBSD/Solaris?

    10. Re:oh boy by Anonymous Coward · · Score: 0

      I haven't seen a "B.S.O.D" (PkZipped English, the acronym, bane of society those are! They tell you we are in a BIG hurry on everything nowadays, including how we talk, lol) in ages!

      In fact, not since Windows 2000 (original release) & a driver I had that needed a pretty massive update, whose name shall remain nameless & as to who wrote it as well.

      I will, however, say it dealt with TV on the PC, but that's it. It's been LONG since patched (nearly 1/2 a decade now)...

      (However, again, it's very long buried, no need to bring up the dead, so-to-speak since the problem's LONG fixed!)

      NOWADAYS?

      Windows Server 2003 SP #1 + hotfix/patches/updates the last 2 years here running it??

      Been excellent, very stable!

      NO problem using WHQL/WDM drivers here from XP/2000 installs as well (one great move MS put into place - making sure their drivers were up-to-par & tested as so as well), & that was really the ONLY bsod generators I had - bum drivers.

      One I have not had in years, as noted/illustrated above...

      APK

      P.S.=> I wouldn't count Microsoft out of anything, even this SuperComputer Cluster version of the OS I am using here in Windows Server 2003 fully updated... I really wouldn't.

      Put it this way - I wouldn't bet on you, betting against them, in anything technical & to be honest? @ technical levels like these in OS design... nothing against you, but they've done a great job with the NT-based OS family, no doubt about that... apk

    11. Re:oh boy by Duhavid · · Score: 2, Funny

      Imagine a Microsoft cluster of those!

      --
      emt 377 emt 4
    12. Re:oh boy by utnow · · Score: 0, Troll

      I'm with you...

      Any closed-minded linux moron who bashes modern MS products clearly hasn't used it in a while. The BSOD is a very rare occurance now-a-days. I can't remember the last time I've seen it... 2 years? give or take

      My machine has been on 24/7 under heavy everyday use in a variety of arenas (gaming, graphics, video/sound editing... etc). Don't even recall the last time I had to reboot...

      Oh! Did I mention virus free? I don't even have any virus protection software installed. I run clamAV every few months just to make sure but I generally don't know what the fuss is about viruses... or spyware. Just don't install the shit. ::shrug::

      Oh! Did I mention that I haven't had to recompile my kernel EVER? And my drivers just have to be double-clicked... done. Sometimes you click 'next' a few times. I'm so glad ALL of the various OS's figured that out a long time ago... it would be rediculous if you still had to access the console and install your various drivers by hand... making sure all of the files are in the right places and such...

      Ooooohhh... ya'll still do that? haha

      It's like trying to drive across the country in a model-t... it might be fun to drive around main-street just to show off that you have it and it's actually running.... but you wouldn't want to go on a cross-country road-trip.

      Linux has always felt like a house of cards to me. It works wonderfully as long as you don't touch it. But if you pull out one card and the whole thing crumbles. At least with windows I don't worry about it suddenly deciding not to boot every time I hit power.

    13. Re:oh boy by Bert64 · · Score: 1

      Actually most of us select the driver (or app) we want to install from a list of packages available for download from the OS vendor.. Then we click install and it downloads and installs...
      No need to trawl round webpages finding drivers, no need to click through license agreements and no need to keep hitting next or rebooting just to install drivers, and no need to fork out money for most of the apps.

      --
      http://spamdecoy.net - free throwaway anonymous email - avoid spam!
    14. Re:oh boy by Scarletdown · · Score: 1

      Actually most of us select the driver (or app) we want to install from a list of packages available for download from the OS vendor.. Then we click install and it downloads and installs...


      Yeah. Linux is pretty easy to use nowadays. I've been using Debian (Sid) for the past couple years, and about the only bit of hardware I have that I can't get to work is an old Parallel port scanner. So, I have that attached to the Windows ME box that I have set up for this and for testing various bits of software I pick up to resell.

      Also, I have never had to compile my own kernel. I suspect that usually is only required if one is trying to get some obscure bit of hardware to work (hardware that most likely doesn't work under the latest incarnations of Windows at all.

      And to twist the GP Poster's claims around...

      Any closed-minded Windows moron who bashes modern Linux distributions clearly hasn't used it in a while (if ever).

      *** This post brought to you courtesy of Mariner of Holland Lager and Grocery Outlet. Now to go have one more before calling it a night. ***
      --
      This space unintentionally left blank.
    15. Re:oh boy by richlv · · Score: 1

      what about electricity consumption ?
      isn't it possible that a difference in maintenance costs (electricity, conditioning, facilities, replacement of out-of-production parts etc) exceeds a cost a newer system ?

      i'm all for complete utilization of older systems, but i don't think home users is a viable market for such solutions.

      --
      Rich
    16. Re:oh boy by hdparm · · Score: 1

      I can't tell whether you're saying this out of your belief or you're trying (very poorly) to troll. Whatever the case, I've got 2 words for you:

      1] FUCK
      2] YOU!

    17. Re:oh boy by Bush+Pig · · Score: 2, Funny

      I guess I'm not the only one who misread it as "Windows Complete Cluster" (so, what's new?)

      --
      What a long, strange trip it's been.
    18. Re:oh boy by Bush+Pig · · Score: 1

      I had a BSOD about 2 days ago (XP Professional, SP 2, yaddah, yaddah, and just standard apps with no weird drivers). It's the first one in a while, but still ... not so stable. It's my work machine (hence bog-standard setup) and gets shut down every night. I wouldn't run it at home, though. Too flakey.

      Linux, OTOH, is rock-solid - certainly no house of cards. I haven't had a problem in years.

      --
      What a long, strange trip it's been.
    19. Re:oh boy by ColonelClaw · · Score: 2, Funny

      can you imagine how utterly superb it would be to simultaneously crash 1000 PCs? it could be a new category in the Guinness Book of Records :D

    20. Re:oh boy by Bert64 · · Score: 1

      Yes, i also have an old scanner (not parallel, something proprietary) that doesn`t work under linux, or any version of windows newer than 95 (it "kinda" works with 98 but is horribly unstable.

      I also have a number of ethernet controllers (sun chipsets) that don`t work with any version of windows, but work perfectly with linux..

      Also many wireless cards, don`t work fully (no monitor mode, arbitrary packet injection or changing your mac address) on windows.

      --
      http://spamdecoy.net - free throwaway anonymous email - avoid spam!
    21. Re:oh boy by wirah · · Score: 0

      Yep, i'm with this guy.

      Linux (Debian) at home, rock solid, installing software is too easy.

      I often go into console purely because I like to do things via console, and not spend all day clicking around a vast array of buttons.

      It has never crashed since I installed it (one and a half years ago), sure the odd app crashes now and then, but never the operating system.

      Shame I have to spend half my working day at work fucking around in windows like I am now. What a waste of time, not to mention slow.

      So keep putting Linux down, stick to your windows. I don't care. I really don't care!

    22. Re:oh boy by Anonymous Coward · · Score: 0

      They should have just called it Windows Clusterfuck Edition.

      They already used up the WINCE acronym.

    23. Re:oh boy by fuzzix · · Score: 1
      Oh, they're controlled. Mostly by opped users in hidden IRC channels.
      Quick question... Why would the users need to be opped to invoke a scripted command?
  2. Little redundant... by HerculesMO · · Score: 1, Redundant

    Windows *is* easier to use than Linux.

    That said, Linux rocks Windows in every other arena.. security, stability, uptime, memory management, etc.

    --
    The price is always right if someone else is paying.
    1. Re:Little redundant... by MightyMartian · · Score: 5, Insightful

      I don't know what the "easy to use" argument precisely means in this arena. The people assembling and managing large clusters to produce supercomputers are in a significantly different league than your average MS-Word user. Not to poo-poo MS or anything, but Gates seems to have forgotten the audience in this case.

      --
      The world's burning. Moped Jesus spotted on I50. Details at 11.
    2. Re:Little redundant... by AuMatar · · Score: 4, Insightful

      In what way? It seems the two are equivalent- double click this icon to start a program. Menu bar over there for a list of all programs. Except Linux has a good text interface as well as the GUI. If anything, that gives the edge to Linux.

      --
      I still have more fans than freaks. WTF is wrong with you people?
    3. Re:Little redundant... by vsprintf · · Score: 1

      It seems the two are equivalent- double click this icon to start a program. . . .

      At one point the KDE GUI seemed to be superior - you (the user) only needed to single-left-click an icon to start a program. Really, why would you wish to single-click an icon and have no reaction? Itchy trigger finger, and you're clicking randomly on the desktop? You use a one-button mouse? I was disappointed when the startup command became a double-click. Why settle for a poor interface just because it's the Windows way? (No doubt there's a way to change KDE's default behavior, and perhaps someone will explain it.)

    4. Re:Little redundant... by drsmithy · · Score: 2, Informative
      At one point the KDE GUI seemed to be superior - you (the user) only needed to single-left-click an icon to start a program. Really, why would you wish to single-click an icon and have no reaction?

      You do get an reaction - the icon is selected.

      You don't always want to open something when you click on it. Sometimes (mostly, I'd wager) you just want to select it for further manipulation.

      Why settle for a poor interface just because it's the Windows way? (No doubt there's a way to change KDE's default behavior, and perhaps someone will explain it.)

      It wasn't changed because "it's the Windows way", it was changed because it makes more sense. Indeed, Windows tried the "single click to launch" way back in 1996-1997 or so, with "Memphis" (later became IE4) and it was soundly canned by pretty much everyone who used it. KDE copying this idea was not one of its smarter moments.

      I'd be interested to see if you can cite any sources for "single-click" to be "better UI" (or even just a well-reasoned argument). Microsoft would have done the reasearch back when they tried it (and Apple probably did some on the same thing way back in the early 80s), but the results are probably not easy to find.

    5. Re:Little redundant... by 10Ghz · · Score: 1
      It wasn't changed because "it's the Windows way", it was changed because it makes more sense.


      But it doesn't make any sense. Windows uses double-click. And just about every single day I see people double-clicking icons on the desktop. And they also double-click the quick-launch-buttons. And they also double-click links in websites.

      Having the GUI use single-click by default merely enforces a consistent behavior across the system. Everything on the desktop would use single-click. But if you use double-click (like Windows and OS X do), then you would have some things using double-click (icons on the desktop and filemanager) while others use single-click (links on websites, quick-launch-buttons). And it gets very confusing for users.
      --
      Lesbian Nazi Hookers Abducted by UFOs and Forced Into Weight Loss Programs - -all next week on Town Talk.
    6. Re:Little redundant... by vsprintf · · Score: 1

      You don't always want to open something when you click on it. Sometimes (mostly, I'd wager) you just want to select it for further manipulation.

      Nonsense. Most often, when you click on an icon, you want to start or open whatever it represents. What possible manipulation is not available while using a two-button mouse and a right-click context menu or drag?

      I'd be interested to see if you can cite any sources for "single-click" to be "better UI" (or even just a well-reasoned argument).

      Sure, I can cite a source - me. As a former (and sometimes still) Windows user, I thought the KDE single-click way was better, and KDE's usage was a lot later than '97. Cite a reputable source that says it's worse (other than your "Memphis" claim which I never heard of).

    7. Re:Little redundant... by drsmithy · · Score: 1

      You do raise a valid point - the double click *is* an unintuitive action. Learners *do* have trouble figuring it out.

      However, once they do, being able to select objects without launching them is an important general behaviour. It's one of those cases where flexibility for the greater userbase wins out over simplicity for the learners.

      Having the GUI use single-click by default merely enforces a consistent behavior across the system.

      Only if you have the assumption that hyperlinks (or quicklaunch icons) == desktop icons. They don't. They look different. They're in a different place. They (typically) abstract different things (a shortcut - "pointer" - vs the actual object.

      Everything on the desktop would use single-click. But if you use double-click (like Windows and OS X do), then you would have some things using double-click (icons on the desktop and filemanager) while others use single-click (links on websites, quick-launch-buttons). And it gets very confusing for users.

      It's confusing until they learn to recognise the different types of UI objects (assuming someone bothers to try and teach them). The UI train wreck of the Dock makes the problem even worse, because the icons on it look just like the icons everywhere else, even though they don't behave the same way at all. At least in Windows there are some visual differences.

      The problem with your reasoning is that you are using hyperlinks and quicklaunch icons as the rule, when they are the exception. The rule is that icons represent objects that need to be double clicked to open, the *exceptions* are elements like quicklaunch buttons and hyperlinks.

    8. Re:Little redundant... by drsmithy · · Score: 1
      What possible manipulation is not available while using a two-button mouse and a right-click context menu or drag?

      Don't move the goalposts.

      Context menus and drag & drop are concepts with notoriously poor discoverability. Not that double-clicking is easy, but once that skill is learned, the "select, then manipulate" model of the desktop metaphor is consistent with it.

      Quicklaunch buttons and the like are the exception - the inconsistency - not the rule.

      Sure, I can cite a source - me.

      This is about as valid as me saying you're wrong.

      As a former (and sometimes still) Windows user, I thought the KDE single-click way was better, and KDE's usage was a lot later than '97.

      I assume you mean "earlier" - and given the KDE project only started in late 1996, it couldn't have been much earlier than 1997 (KDE 1.0 was released 1998).

      Cite a reputable source that says it's worse (other than your "Memphis" claim which I never heard of).

      Sorry, I got my code names mixed up:

      "Nashville" was the one I was thinking of - it was the name used for IE4 while it was in development (these development versions of IE4 and Windows were also bandied around warez groups as "Windows 96" and, later, "Windows 97"). This was the first integration of IE into Windows - particularly the UI - when they were trying to make everything look and feel like a web browser. So this was when the "single click launch" first reared its ugly head. A quick Google shows this page mentions it:

      Internet conventions will be used throughout, so that a single mouse-click on an object--be it program shortcut, document shortcut or a filename in the file window--will start the originating application and load the file.

      My memory is a bit hazy, but the first versions of "Nashville" were surfacing late 1997. That's when the first examples of "single click launching" on Windows would have been around.

      I don't have any HCI research on hand about single-click launching, and I'm not going to waste hours researching something I really don't care about. However, I do remember quite well how unpopular it was during the IE4 betas and release - enough to have it changed to "off" as default by the time Windows 98 shipped. The fact that Apple has never tried it would also suggest their UI research indicated it was poorly received as well.

      It's been an easily accessible option in Windows since IE4 was released and in bits of beta software since 1997 and talked about since 1996. If you can dig up either of the two official IE4 betas it's definitely on by default in them, and I'm pretty sure it was on by default in the initial release of IE4 as well.

    9. Re:Little redundant... by 10Ghz · · Score: 1
      However, once they do, being able to select objects without launching them is an important general behaviour.


      I haven't had any problems in selecting icons without launching them, even when using single-click.

      Only if you have the assumption that hyperlinks (or quicklaunch icons) == desktop icons. They don't. They look different. They're in a different place. They (typically) abstract different things (a shortcut - "pointer" - vs the actual object.


      You are right. But users don't care. They are "opening folders" and "opening websites" and "opening apps". And they do that by double-clicking. To them that is the way you open them, they don't really see any difference between quick-launch icons, links and icons.

      The problem with your reasoning is that you are using hyperlinks and quicklaunch icons as the rule, when they are the exception. The rule is that icons represent objects that need to be double clicked to open, the *exceptions* are elements like quicklaunch buttons and hyperlinks.


      But these day hyperlinks are NOT the exception. People propably use links more than they use objects in filemanager and desktop. And in both cases, the user "opens" the object, but for some reason it's done differently in websites and in the desktop. And to the user, that gets very confusing.
      --
      Lesbian Nazi Hookers Abducted by UFOs and Forced Into Weight Loss Programs - -all next week on Town Talk.
    10. Re:Little redundant... by vsprintf · · Score: 1

      Don't move the goalposts.

      Context menus and drag & drop are concepts with notoriously poor discoverability. Not that double-clicking is easy, but once that skill is learned, the "select, then manipulate" model of the desktop metaphor is consistent with it.

      How so? Dragging is standard mouse usage. Right-click context menus are standard behavior for Windows and Linux desktops. As you point out yourself, double-clicking is hardly intuitive - it is simply the Windows way.

      Quicklaunch buttons and the like are the exception - the inconsistency - not the rule.

      That seems to support my argument more than yours.

      This is about as valid as me saying you're wrong.

      I have used Windows since version 1.0. If anyone should be resistant to change, it should be me. However, I liked KDE's single-click activation. For icons, it makes more sense. It is not a matter of right or wrong - it's the way it is.

      I assume you mean "earlier" - and given the KDE project only started in late 1996, it couldn't have been much earlier than 1997 (KDE 1.0 was released 1998).

      No, I said later, and I meant later. You were the one saying how all this had been proven by '97. It was only a few years ago that KDE still had single-click activation. Apparently, the KDE developers (and I) still thought it was the better method.

      The fact that Apple has never tried it would also suggest their UI research indicated it was poorly received as well.

      That would more likely be a result of Apple's belief that mouse users have only one finger and that Apple systems ship with a one-button mouse. Okay, your whole argument seems to be that MS received some negative feedback from beta testers. Well, some people just hate change. To me, single-click activation for icons (whether they are on the desktop or in the tray) seems more intuitive, and it's certainly more efficient. There is rarely a reason to "select" an icon, and I've yet to see one that couldn't be accomplished in another fashion.

      Just think how many clicks could be saved world-wide if single-clicking icons became the standard. All those saved fractional seconds could easily result in man-years of increased productivity. And then there's the reduction in repetitive stress injuries. :)

    11. Re:Little redundant... by drsmithy · · Score: 1
      How so?

      Because neither are particularly intuitive if you've never used a mouse before.

      Dragging is standard mouse usage.

      Of course it is - once you've figured out how to do it. Which aspect of "discoverability" is difficult to understand ?

      Right-click context menus are standard behavior for Windows and Linux desktops.

      That doesn't mean they're easy to find.

      The Windows UI guidelines specifically state that context menus are to be used as accelerators only, and that no functionality should be only exposed via a context menu. MacOS took some time to embrace context menus and even today, they're practically nonexistant. I assume - although I haven't read them - that KDE's and GNOME's UI guidelines have similar specifications.

      The reason for this, is because context menus have - as I said - notoriously poor "discoverability".

      As you point out yourself, double-clicking is hardly intuitive - it is simply the Windows way.

      Stop using this as an excuse to exercise your bias. Double clicking is "the way" used (both presently and in the past) of the vast majority of every computer GUI ever made. It's not even *remotely* "the Windows way".

      For icons, it makes more sense. It is not a matter of right or wrong - it's the way it is.

      If you're trying to make a general statement, then no, it's not - as supported by my anecdotal evidence of personally preferring a double-click mechanism.

      No, I said later, and I meant later. You were the one saying how all this had been proven by '97.

      I didn't say anything had been "proven" by 1997. I said it had been tried out in Windows around 1997 and been soundly rejected by end users.

      It was only a few years ago that KDE still had single-click activation. Apparently, the KDE developers (and I) still thought it was the better method.

      OSS developers aren't exactly renowned for their HCI research. Even moreso back in that timeframe. UI features in OSS software tend to be a) what the developers want and/or b) what the technically-adept userbase wants.

      That would more likely be a result of Apple's belief that mouse users have only one finger and that Apple systems ship with a one-button mouse.

      If anything, Apple would be the *first* to start using a procedure that makes the UI easier.

      Okay, your whole argument seems to be that MS received some negative feedback from beta testers.

      No, my arguments are that Microsoft received negative feedback from just about everyone - beta testers, reviewers, end users (remember, this happened in the 1997 - 98 timeframe when IE was busily wiping the floor with netscape and doubling its userbase every couple of months), that the company known for careful HCI reserach and good UI and that usage of the single-click methodology - despite being an option in Windows for ~8 years - is extremely rare, all suggest that the vast majority of people do *not* think it is a better way.

      There are also UI consistency problems. For example, if you can't single-click an icon without launching it, and right-clicking opens a context menu, how do you consistently deal with trying to select multiple non-contiguous icons (ie: Ctrl+clicking in Windows) ?

      Well, some people just hate change. To me, single-click activation for icons (whether they are on the desktop or in the tray) seems more intuitive, and it's certainly more efficient. There is rarely a reason to "select" an icon, and I've yet to see one that couldn't be accomplished in another fashion.

      Well, that's why the _option_ exists - for people who prefer it.

      Just think how many clicks could be saved world-wide if single-clicking icons became the standard. All those saved fractional seconds could easily result in man-years of increased productivity. And then there's the reduction in repetitive stress injuries. :)

      Ah, now this reminds of the standard Mac-zealot "how many clicks does it take to do this on Windows" "benchmarks" :).

  3. Confused by Henry+V+.009 · · Score: 5, Funny

    "easier to use than Linux"

    Yes? Where is the part about the high hopes for this operating system?

    1. Re:Confused by rco3 · · Score: 2
      "easier to use than Linux"

      Yes? Where is the part about the high hopes for this operating system?

      That would be the "as powerful" part.
      --

      Ce n'est pas un vrai mouvement de robot!
    2. Re:Confused by jcr · · Score: 1

      Hope springs eternal!

      -jcr

      --
      The only title of honor that a tyrant can grant is "Enemy of the State."
    3. Re:Confused by sedyn · · Score: 1

      "Where is the part about the high hopes for this operating system?"

      Well, I'm kinda excited about the thought of having more computers crash faster. Providing more in less time should surely be desirable by the public.

      As a side note I wonder if there is a "Quickest computer crash" Guiness World Record? (well, for a system not in development)

      --
      Am I open minded towards open source, or closed minded towards closed source?
    4. Re:Confused by Quevar · · Score: 2, Insightful
      "Yes? Where is the part about the high hopes for this operating system?"
      The high hopes for the OS are back when the name suggests it was supposed to have shipped. Why is it called: "Windows Compute Cluster Server 2003" and they are only now (in 2005) getting a second Beta out? So, is this going to be two years out of date (or more) by the time it ships?
    5. Re:Confused by vettemph · · Score: 1

      Yep, Linux will always be almost good enough to use on your desktop, Windows will always be almost good enough to safely connect to a network (without loads of third party life support software).
        Someday (soon) linux will be good enough for the desktop. (it's been good enough for me for over five years now and I ain't no programmer.)

      --
      The government which is strong enough to protect you from everything is strong enough to take everything from you.
    6. Re:Confused by KarmaMB84 · · Score: 1

      The name is probably related to it being part of the Windows 2003 line of operating systems...

    7. Re:Confused by Scarletdown · · Score: 1

      As a side note I wonder if there is a "Quickest computer crash" Guiness World Record?


      That honor would go to my old TI-99/4a. Crashing as soon as the power switch was turned on because the Parsec game cartridge wasn't quite inserted all the way. :D

      --
      This space unintentionally left blank.
  4. BSODs and Windows EULA's by einhverfr · · Score: 0, Redundant

    Imagine a Beowulf Cluster of BSOD's and Windows EULA's.

    No thanks. I don't think they will get very far here, certainly not in the main computing nodes (maybe for peripheral nodes, but not in the heart of the cluster).

    --

    LedgerSMB: Open source Accounting/ERP
    1. Re:BSODs and Windows EULA's by middlemen · · Score: 1, Insightful

      Imagine rebooting the complete cluster after installing software !!!

    2. Re:BSODs and Windows EULA's by Anonymous Coward · · Score: 0


      Node 1: Please insert your original Genuine Windows CD-ROM and enter your Product Key now.
      Node 2: Please insert your original Genuine Windows CD-ROM and enter your Product Key now.
      Node 3: Please insert your original Genuine Windows CD-ROM and enter your Product Key now.


      ... sigh...

    3. Re:BSODs and Windows EULA's by einhverfr · · Score: 1

      It was occuring to me that if you connected everyhting in a hypercube, you might be able to get by with Windows XP Home Edition and SFU 3.5.....

      --

      LedgerSMB: Open source Accounting/ERP
    4. Re:BSODs and Windows EULA's by pv2b · · Score: 1

      Sysprep.

    5. Re:BSODs and Windows EULA's by Anonymous Coward · · Score: 0

      Is the cluster doing work for you right now? Sorry, Automatic Updates had to reboot the entire cluster because it installed a critical update to Internet Explorer.

  5. I wouldn't trust Bill Gates by Anonymous Coward · · Score: 2, Funny

    I think he might own some Microsoft stock.

  6. First thing one associates with that name... by Anonymous Coward · · Score: 5, Funny

    ... well for me, it was clusterfuck.

    1. Re:First thing one associates with that name... by Anonymous Coward · · Score: 3, Funny

      Microsoft Windows Complete Clusterfuck? Why, that's not a new version at all!

    2. Re:First thing one associates with that name... by Hymer · · Score: 1

      "... well for me, it was clusterfuck."
      You've got everything wrong, dude:
      Clusterfuck is good, it is like a Roman orgy. Just imagine, you and a lot of naked beautiful women.
      I think that you were talking about something called "clustershit"...
      --
      I suppose I am sick, I do however not want to be cured... ;-)

    3. Re:First thing one associates with that name... by Dr+Floppy · · Score: 1

      exactamundo. This is totally stupid for MS to try. Why would anyone want to use Windows when Linux, UNIX/MacOSX do the job and are sure to be less buggy.

    4. Re:First thing one associates with that name... by Blkdeath · · Score: 1
      You've got everything wrong, dude:
      Clusterfuck is good, it is like a Roman orgy.

      {whoosh} would be the sound of John Wayne's line going over your head.

      Otherwise used extensively by the military. Not surprisingly, Wikipedia is wrong. :)

      --
      BD Phone Home!

      Shameless plug. Like you weren't expecting it.

    5. Re:First thing one associates with that name... by Cyberia · · Score: 1

      Your half right...

      When archeologists discover a 1,000,000 year old windows 2003 compute cluster they will undoubtedly call it the Clusterfckasaurus.

    6. Re:First thing one associates with that name... by Anonymous Coward · · Score: 0

      tacolicker? On Slashdot? That takes karma whoring to a new low. Ewww.

    7. Re:First thing one associates with that name... by Anonymous Coward · · Score: 0

      Aye, nothing new here. These have been around for years, easy to manage too, operator just sits in an IRC channel and issues commands from there. Microsoft just wants to embrace and extend.

    8. Re:First thing one associates with that name... by Bert64 · · Score: 1

      Because of marketting..
      It doesn`t matter how massively inferior and more expensive windows is, microsoft will market it to the non technical people in higher positions and they will commit to it without consulting anyone more knowlegeable, leaving those who know what they`re doing with a huge mess to clean up.

      --
      http://spamdecoy.net - free throwaway anonymous email - avoid spam!
    9. Re:First thing one associates with that name... by Bush+Pig · · Score: 1

      ... and it will still be rebooting itself from a software installation.

      --
      What a long, strange trip it's been.
  7. How much? by rkh17 · · Score: 1

    But, how much will it cost?

    1. Re:How much? by CastrTroy · · Score: 3, Informative

      Exactly. With microsoft usually charging per processor, the cost of building a super computer could go through the roof. Not to mention that they will probably build in limitations like maximum 4 processors, and to use more, you'll have to buy the enterprise edition, and spend 3 times as much per processor.

      --

      Anthropic principle: We see the universe the way it is because if it were different we would not be here to see it.
    2. Re:How much? by dattaway · · Score: 2, Funny

      Unfortunately, it takes a supercomputer to calculate that answer.

    3. Re:How much? by MightyMartian · · Score: 4, Insightful

      It's hard to see how MS will gain much market penetration in this area unless it substantially alters its licensing. As I said in another post, "ease of use" isn't exactly going to be a big selling point for the guys putting these kinds of computers together, and OS licensing costs of $0.00 is hard to beat.

      --
      The world's burning. Moped Jesus spotted on I50. Details at 11.
    4. Re:How much? by schon · · Score: 4, Interesting

      "ease of use" isn't exactly going to be a big selling point for the guys putting these kinds of computers together

      Excuse me while I find my tinfoil hat.... .. OK, that's better:

      What if MS is doing this so that it can strongarm universities and research institutions? Something like going to the bean counters and saying "hey, we have this great new OS for supercomputers - we'll give you a reduced rate on it, *AND* a reduced rate on the licenses for the rest of your desktops, if you just agree to kick that smelly, communistic, viral, legally-dubious Linux off your clusters! (Did we mention that if you're using Linux, you might have to give up all your precioussss IP?)"

    5. Re:How much? by Stevyn · · Score: 1

      Microsoft probably invested a good chunk of change into this so it makes me wonder..are they trying to turn a profit on this? If they are, the market is probably small and they'll have to charge a lot per license to make any real money on it. Or, this is more or less to build up their image to impress those wishing to buy their current enterprise software for other large tasks. Bottom line, this is probably more for show than to directly turn a profit.

    6. Re:How much? by Sax+Maniac · · Score: 4, Insightful
      MS is not stupid. They know exactly what supercomputer owners are paying in licenses as part of the total cost (they're not all Linux), and I'm sure they'd steeply discount the license to get into the space they want.

      They bring very little value to the back end. A typical distributed app is not a rich UI client that needs lots of Windows APIs to play DRM'd movies, so Windows has no advantage there. It's a C, C++, or Fortan (mixed, even) job running MPI over some specialized interconnect hardware.

      You also need a good parallel file system which I'm willing to go out on a limb and say that CIFS is probably not the optimal choice - any real system will probably be using a dedicated filer.

      However, their strong suit in this space are tools like IDEs. If they can convince folks that using Windows as a front-end to development, then they can make some good inroads.

      Right now the supercomputing folks are starting to get interested in Eclipse, and they're trying to head that off, not to mention small ISV's like us.

      --
      I can explanate how to administrate your network. You must configurate and segmentate it, so it can computate.
    7. Re:How much? by MightyMartian · · Score: 4, Insightful
      And that means that those universities are going to have rewrite their custom apps at great expense. I can't see it. I simply can't see how MS can reasonably strong arm anyone, corporate, governmental or university, into taking on a supercomputer variant of Windows.

      And even if MS were so compelled, what really is in it for Redmond? It's probably the smallest market in the world, with a customer base measured in the thousands, and one that already has access to either operating systems with a long multicomputing heritage or to clustered Linux systems that, for the kinds of guys that set up and maintain supercomputer clusters, offers no unreasonable difficulties as far as usability.

      I think what we have here is Gates' inferiority complex towards Linux. He desperately wants to have MS in every market, and can't stand the thought of that open source demon increasingly being utilized, so he'll waste money on a product that, even if it were to be a success within that marketplace, would provide an outrageously small amount of revenue.

      I suppose MS has money to burn, but if I had that much money to piss away, I'd try to take over the pocket calculator or security alarm markets.

      --
      The world's burning. Moped Jesus spotted on I50. Details at 11.
    8. Re:How much? by tolkienfan · · Score: 1

      I agree, plus Windows will never be flexible enough. Never. Ever. Period.

    9. Re:How much? by InvalidError · · Score: 1

      I really am looking forward to Windows Vista for Supercomputers where each node will require a DX9-compliant graphics sub-system.

      The only application-level interface supercomputing nodes need is very fast MPI. With many supercomputers running often disk-less ultra-lightweight BSD or Linux kernels on computing nodes to leave as much RAM and CPU resources to the software as possible, it would make no sense to have all of Windows' AV-centric overhead.

      How much will it cost? Too much performance-wise, too much hardware-wise and too much license-wise.

      Win-CSS makes no sense... unless Bill wants to have an entry in the top-500 and is willing to give big scholarships and subventions to offset the costs of renting and deploying Win-CCS licenses.

    10. Re:How much? by rufty_tufty · · Score: 1

      "I'd try to take over the pocket calculator or security alarm markets."

      Didn't they already try that with embedded windows - sorry, but this is the best link I could find:
      http://msdn.microsoft.com/embedded/

      --
      "The weirdest thing about a mind, is that every answer that you find, is the basis of a brand new cliche" -
    11. Re:How much? by Wizarth · · Score: 1
      I think what we have here is Gates' inferiority complex towards Linux. He desperately wants to have MS in every market, and can't stand the thought of that open source demon increasingly being utilized
      Demon? I thought FreeBSD was dead?
    12. Re:How much? by Ninjy · · Score: 2, Interesting

      Windows on the alarm clock? Sign me up for one of those.

      Finally I'll be able to use the "my alarm clock didn't work" excuse a few times a week. :)

    13. Re:How much? by sumdumass · · Score: 1

      It isn't neccesarily about the universities and rewriting code. Although microsoft might go ahead and do it for them under the conditions they switch or use thier emulators.

      What this is about is some MCSE or PHB needing extra computing power and microsoft selling it. It is about using/extending thier new multiple core processor support for more then just gaming machines. My bet is on microsoft actualy creating a new market place with it's newer OSes using more processing power and needing extra umph for aplications to run then ever before. Soon companies making processor intensive aplications will start supporting multiple cores and then it isn't much of a step futher to support a cluster (cores on different computers) of cores.

      It won't be long before Microsoft would be able to say "here is free development tools and here is a language/platform most trained monkeys can use. It fits in with existing tools and technoligy, spend your money here". Then it would be trivial to back it up with some return on investment study or modified TCO report that ilistrates MS wrenches get paid less then *nix admins. Bundling will soon ensure They have a new market with a foothold in the old.

    14. Re:How much? by Anonymous Coward · · Score: 0

      But..did you know that redhat uses the same stratagem and makes you wait exactly (n) seconds
      in their WS releases if you have more..while complaining that they don't 'support' it, and
      the logical correlative is that you are a walking pile of dung for not buying their server
      product to accentuate the goodness of their borkenness. (What vendor backports NPTL to the
      2.4 tree without realizing that sh!t goes south quick?)

    15. Re:How much? by EraserMouseMan · · Score: 1

      Setting up a supercomputer is not a simple matter of installing the OS on a box. With Linux the OS may be free. But I guarantee you that the professionals who are designing and implementing the solution aren't doing it for free. So, a Linux supercomputer is not $0.00 and neither will a Windows one be.

      It would be interesting to see a pie chart of the costs involved with a supercomputer. I bet the hardware is a giant slice of the pie. And I bet that the OS licenses are next to nothing regardless of which OS you use.

    16. Re:How much? by Anonymous Coward · · Score: 0

      Yes asshat, some of them are *BSD.

    17. Re:How much? by fitten · · Score: 1

      And that means that those universities are going to have rewrite their custom apps at great expense.

      Doubtful... many (I would almost say "most") HPC apps are written in MPI and there are a number of MPI implementations on Windows. I've personally used one and the "rewrite" consisted of building a project in MSVC and compiling. I think it took about five minutes. For those not afraid of Windows, it isn't a difficult task.

    18. Re:How much? by mcrbids · · Score: 1

      And even if MS were so compelled, what really is in it for Redmond? It's probably the smallest market in the world, with a customer base measured in the thousands, and one that already has access to either operating systems with a long multicomputing heritage or to clustered Linux systems that, for the kinds of guys that set up and maintain supercomputer clusters, offers no unreasonable difficulties as far as usability.

      Ever watch NASCAR? Cars going around in circles with signs plastered all over them: "PENZOIL", "SHELL", "BP", etc. It's all about getting your name out.

      For Microsoft, the Supercomputer 500 is a big pissing ground. They've always been about the little potatoes of the personal computer - can they compete with the "big boys" in the top 500? And, can they do so publicly so they can say: "use the same O/S used by NNN to make a top 10 supercomputer!"

      Where, by the way, they're getting dusted by the same O/S dusting them in commodity servers, and giving them a good run for their money in the embedded/cell-phone space.

      So, they face a competitor that operates smoothly in the very large space, the very small space, and in the very profitable midrange, too. Can they take this lying down?

      --
      I have no problem with your religion until you decide it's reason to deprive others of the truth.
    19. Re:How much? by Anonymous Coward · · Score: 0

      Oh, I can respond to that fairly quickly. Here goes: The Prof. to the beancounter: you can put any kind of shit operating system on your desktops, but on the supercomputers, get the fuck away. Microsofts systems are 1. not as powerful as Linux (read they can't get it up, they are impotent, slow as shit, unstable, insecure) and 2. Linux scaled to over 1000 machines more than 10 years ago. Windblows clusterfuck edition can at best (potentially) get to 64 machines. At some point, being able to publish and be considered an un-fucked university means more then getting a cut-rate price on crapware. Here's another proof-in-the-pudding concept: Windblows clusterfuck edition came out 3 months ago. In that time, Linux has gained on the top500.org list (to 74%+ of all machines, including 9 out of the top 10). Microsoft has exactly 0 machines in the top500 (there was 1 machine a year ago, but Microsoft doesn't have enough money to buy enough processors to keep it there). I guess the real problem is that when talking about supercomputers, they aren't run by pointy haired bosses or less-than-newbie home users who keep running back to the store with 'I think I got a virus'. The people who put these machines together have a clue. Microsoft marketing comes off soundling exactly like the bullshit it is. 64 processors is about 5% of what it takes to make the bottom of the list. On the low end, 1000 processors are needed. With IBM's BlueGene systems, 130,000 processors get you the top of the list (oh, and it runs Linux). There is a long way between 130,000 and 64. Conclusioin: Mickeysoft clusterfuck edition: This dog don't hunt.

    20. Re:How much? by seweso · · Score: 0

      Nice website you have....without javascript!

    21. Re:How much? by Anonymous Coward · · Score: 0

      OK, people who build supercomputers are not stupid either. On the top500 list, 74% of the worlds fastest 500 run Linux. They know that Microsoft's systems are not as efficient (Microsoft has recently admitted this). Not just marginally less efficient, but radically less efficient. Steep discounts don't amount to much when the difference in price means not 'getting there'. Why would anyone spend 15 million dollars on hardware, and save 20% on operating system software that makes the hardware look like it was made 30 years ago? People want the best bang for the buck here. Microsoft's stuff has not been the best for a long time. Everyone knows they aren't dedicated to quality. The people who build these systems have a clear goal in mind. Microsoft doesn't 'kinda get them there'. It doesn't get them there at all. At best, Microsoft's system runs on 64 processors. Linux runs on (at least) 130,000. You can build a 'toy' supercomputer if you have money to burn, but the thing is: if you don't know the difference or don't know any better, then you don't know what to do with a supercomputer anyway. The other point is: why pay 20% less for a lesser system when you can get a better one for free (and the nonsense about having to share your data gets squelched very quickly too, that little 'what-if' might wash with the mindless-simpleton set, this crowd will hurd you out of their machine-room on the toe of their boot if you try to push that crap.

    22. Re:How much? by Bazzalisk · · Score: 1
      I work at a major British college, and a lot of my friends work at another one (no indications of which, of course).

      Every year the adminstrata down in the BLNAK building declare that all computers in the college must run Microsoft Windows - and every year the Maths, Physics, and Aeronautical Engineering departments send a small delegation to explain to Sir. BLANK ' cronies that this is not an option, and they will continue to use Linux, Solaris, and Irix (the computing department just ignores the dictat).

      Sadly I can see this somewhat weakening their case.

      --
      James P. Barrett
    23. Re:How much? by jesterpilot · · Score: 1

      So the university management talks to the scientists about this (they sometimes do!). The scientists reply something like "do WHAT with our precious cluster...?!?", university management spends a short time thinking about core business. University management stumbles upon some desktoplinux company, connected with the computer science department of same university which happenes to host a mirror for this company. Company offers better deal. Some scientists mention they're already using it. It turns out 80% of the systems in the labs runs some form of linux with custom apps the researchers won't leave alone in case of a nuclear assault. University management thinks again about core business, for a very short time now.

      Gates will not risk this scenario.

      --
      Trust me, I work for the government.
    24. Re:How much? by ookaze · · Score: 1

      MS is not stupid. They know exactly what supercomputer owners are paying in licenses as part of the total cost (they're not all Linux), and I'm sure they'd steeply discount the license to get into the space they want.

      You can be sure, you can state things, that does not make them true. I still don't understand what they are trying to acheive here. This move reinforces my belief that MS really think people love to use their OS.

      They bring very little value to the back end.

      Actually they bring none.

      Windows has no advantage there.

      So we agree they bring no value.

      It's a C, C++, or Fortan (mixed, even) job running MPI over some specialized interconnect hardware.

      Remember what you said here ...

      You also need a good parallel file system

      I don't think so, perhaps for some niche apps. Most of the data travels through IPC and highly efficient and fast communication mechanisms, not files. Files are for the end results on the master system.

      However, their strong suit in this space are tools like IDEs. If they can convince folks that using Windows as a front-end to development, then they can make some good inroads.

      Remember what you said earlier ? I don't understand this statement of yours.
      I don't know of any MS IDE that works well for developing Unix/Linux apps.
      IIRC MS IDE (till the latest versions) is only powerful with MS APIs, not at all with FORTRAN or generic compliant C and C++, POSIX, or even MPI.
      Of course, the goal of a cluster is not to have a cool IDE, it is to provide the fastest possible accurate results. So I don't see what their cluster OS is for. Don't tell me they made an OS just to run an IDE !? That looks pretty stupid to me, despite what you believe about MS.
      As soon as you try to talk of them replacing the cluster OS with MS one, it's already a stupid solution, in terms of cost, reliability, efficiency. They just can't beat Linux on cost, no need to even talk about the other terms. And old Unixes still used in these clusters will be replaced with Linux, not Windows.
      I guess Windows can pay some first sites to prove their OS (for now it's still beta), but in the end of the day, people putting clusters in place need results, they can't put up with BSOD or locks or poor perfs.

      Right now the supercomputing folks are starting to get interested in Eclipse, and they're trying to head that off, not to mention small ISV's like us.

      In case you did not remember, Eclipse runs on these OSes used for clusters. No MS app does, and I'd even dare to say none of their OS does (except an hypothetical unknown one).

    25. Re:How much? by Anonymous Coward · · Score: 0

      Doesn't quite work that way. The bean counters have about zero control over what individual researchers do with their grant money, since the grants are given to the researchers, and not the universities. People can and do get fed up with one institution and move to another, with their funding, and if they've got a lot of grants, they can often name their terms. And it's the individual researchers and their grants that build the clusters and supercomputers. Like my lab.

      Maybe strong-arming government research labs is more likely, I dunno.

    26. Re:How much? by pzs · · Score: 1
      Maybe they're doing it for the prestige. The same reason VW want to pour into the Bugati Veyron, or Renault do Formula 1.

      Peter

    27. Re:How much? by thealsir · · Score: 1

      My old university was already forced into that. A Windows cluster. A friend of mine who helped put the thing together regarded it as "regrettable but necessary." If this isn't m$ hegemony, I don't know what is.

      --
      Do not downmod posts "overrated" simply because you disagree with them.
    28. Re:How much? by Sax+Maniac · · Score: 1
      Listen, I've logged into personally and run applications on these supercomputers. The guy down the office is logged into BlueGene right now, and I talk with the folks who built and use these machines all the time.

      A lot of development and debugging for hi-scale actually happens at small scale. They'll get their app working at 2 or 16 or 32 nodes, debug the hell out of it, and then book time on the behemoth to do the real work at a few thousand. It's very expensive to drop a buggy app on a 2,000 node run.

      You have to realize that the licensing fees are a factor, but NOT a huge factor in evaluating the cost for a such system. If that was the case they would never have used propietary systems like AIX in the first place.

      This is not to disparage Linux. I love Linux on supercomputers, it's much more pleasant to work with than other things like AIX.

      I'm not saying Windows is a good idea on a cluster, just that MS is not stupid, and ignore them at your own peril. There are also a market for not-so-big clusters, too.

      The fact that supercomputers are moving to x86 should make us Linux fans more wary. MS could say something like You've already done the hard work porting over to x86. Why not run Windows on your development cluster and front end and use these wonderful UI tools that you love?

      --
      I can explanate how to administrate your network. You must configurate and segmentate it, so it can computate.
    29. Re:How much? by Sax+Maniac · · Score: 1
      I don't think so, perhaps for some niche apps. Most of the data travels through IPC and highly efficient and fast communication mechanisms, not files. Files are for the end results on the master system.

      Have you ever looked at the inside of an app like this? Or are you just guessing?

      You don't use files for IPC, you use it to store giant input data sets and output results.

      Where else are you going to put the result of a 2,000 node computation that produces a few terabytes of data? Span it across 2,000 hard drives?

      Or do you think the answer is just going to be a few bytes long like "42"?

      --
      I can explanate how to administrate your network. You must configurate and segmentate it, so it can computate.
    30. Re:How much? by Scarletdown · · Score: 1
      Exactly. With microsoft usually charging per processor, the cost of building a super computer could go through the roof. Not to mention that they will probably build in limitations like maximum 4 processors, and to use more, you'll have to buy the enterprise edition, and spend 3 times as much per processor.
      And I can see them press it even further by requiring a seperate license for each machine in the cluster to connect to each other, sort of like a CAL (if I'm understanding CALs correctly).
      --
      This space unintentionally left blank.
  8. Maybe Linux... by NardofDoom · · Score: 4, Informative
    But not OS X

    And I can bet it won't be included with their client systems.

    --
    You have two hands and one brain, so always code twice as much as you think!
    1. Re:Maybe Linux... by Coryoth · · Score: 1

      If you want an "easy to use" linux cluster try Scyld, it's pretty much plug and go integrated linux cluster solution out of the box. Then again, that's what you would expect from a distribution tailored explicitly for clustering and HPC solutions.

      Jedidiah.

    2. Re:Maybe Linux... by tskirvin · · Score: 1

      Clustermatic is Scyld-but-free.

      Oh, and for kicks, check out our cluster-building workshop.

    3. Re:Maybe Linux... by NitsujTPU · · Score: 1

      Yes, the supercomputer software probably won't be included with the software for PCs... imagine that.

    4. Re:Maybe Linux... by NardofDoom · · Score: 1
      Okay, I'm going to assume you didn't read the link, and you're not just a troll

      Apple has this technology called XGrid, which lets you perform cluster computing tasks very easily. The client is installed with OS X, and takes zero setup beyond checking off a box in the system preferences and making sure your firewall is open on a specific port. Now, whenever you go onto a network, your Mac searches for an XGrid project to participate in, requests a work unit, and starts using those spare processor cycles. Or you can connect to an existing XGrid Server over the Internet.

      Apple also makes a rack-mounted computer specifically designed for clustering. This has been used to great effect by Virginia Tech with their SystemX.

      So, yeah, their supercomputing software is included with their PCs and notebooks, for a very good reason.

      --
      You have two hands and one brain, so always code twice as much as you think!
  9. An Xbox Cluster? by donweel · · Score: 1

    Can he be serious?

    --
    Many a long talk since then I have had with the man in the moon; he had my confidence on the voyage. Joshua Slocum
    1. Re:An Xbox Cluster? by Azraael · · Score: 1
  10. Why not by Anonymous Coward · · Score: 0

    Why crashing a windows server when you can crash a complete cluster. Great idea!

  11. Silly by Anonymous Coward · · Score: 0

    As someone who works in the high-performance computer industry I find this laughable. It's one thing to argue the merits of Windows on the desktop enviroment...but on clusters...are you kidding me? This is a non-story.

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

      As someone who works in the red light district of Amsterdam, I find this comment laughable. It's one thing to argue the merits of homosexual Macintosh "Oh-ess-ecchs" in the bathhouse ... but on real computers ...are you cornholing me? This is a non-story.

  12. And I compete in the Olympic 100-meter dash by Anonymous Coward · · Score: 0

    I get my ass kicked, but I compete.

  13. Wake up, Bill by Fished · · Score: 5, Insightful

    Supercomputers aren't about "Ease of use." They're about speed per dollar. When WCC can beat Linux on price/performance, then people will stand up and take notice. Not before.

    --
    "He who would learn astronomy, and other recondite arts, let him go elsewhere. " -- John Calvin, commenting on Genesis 1
    1. Re:Wake up, Bill by ma_luen · · Score: 4, Interesting

      Actually the importance of the raw performance of the machine is on the decline. More emphasis is being placed on the idea that super computers are only useful in the sense that they help researchers solve problems. So there is growing interest in the notion of "time to solution" as a combination of ease of programming for, ease of using, and of course running a data set on the machine.

      Mark

    2. Re:Wake up, Bill by dtfinch · · Score: 4, Funny

      I'm pretty sure supercomputers are about performance.

    3. Re:Wake up, Bill by schon · · Score: 4, Insightful

      It's not even that - clusters aren't about shiny desktops, they're about custom-written apps.

      MS Excels at letting the computer illiterate do whatever MS has envisioned they might. You wanna connect to the internet via DSL? No problem. You wanna write an email to Grandma? No problem. You wanna do something that MS hasn't thought of yet? BIG Problem.

      Pretty much *anything* you do with a cluster is gonna be custom, "MS hasn't thought of this" stuff, and so will be harder to use, not easier.

      Why do I get the feeling that it will have a little version of Clippy asking "It looks like you're doing climatology variance research. Would you like me to help you model your data?"

    4. Re:Wake up, Bill by interiot · · Score: 1

      I can't find the quote, but Google had a comment about how Linux allows Google to customize and control ANYTHING in pursuit of higher performance means a great deal to them. Google shouldn't have to wait on Microsoft to implement performance tweaks for them in the kernel.

    5. Re:Wake up, Bill by Otter · · Score: 4, Informative
      Supercomputers aren't about "Ease of use."

      Obviously the comment about "easier to use" is inane when talking about supercomputers, but that quote was invented by the submitter. What the director of the HPC unit (not Gates) actually said was "...easier to integrate into what they are already doing".

    6. Re:Wake up, Bill by GreyWolf3000 · · Score: 2, Insightful

      But I think we all know that MS will be able to spend some of its cash reserves to produce a more scalable and industry-accepted solution than Linux with li[tt]le effort. Just like every other market Linux and Windows have competed in.

      --
      Slashdot: Where people pretend to be twice as smart as they really are by behaving like children.
    7. Re:Wake up, Bill by jpetts · · Score: 4, Insightful

      That's right. So, in quantum chemistry for example, you can run more accurate calculations with a faster computer, and thereby, it is hoped that more accurate, or at least applicable results will be obtained. Therefore, the "time to solution" is reduced. The same applies to CFD, bioinformatics, whatever. There will always be classes of problems where more compute power will give you more data, and possibly insights. Ergo, a faster machine, in the absence of any other factor, will be better for some classes of research.

      --
      Call me old fashioned, but I like a dump to be as memorable as it is devastating - Bender
    8. Re:Wake up, Bill by ma_luen · · Score: 2, Informative

      Yes but the issue is all the performance doesn't matter if your researchers aren't using it to solve problems. See http://www.darpa.mil/ipto/programs/hpcs/ for more info on this. This is the big HPC push that IBM, Cray, and SUN are participating in. Also a company that I think is kinda cool http://www.orionmulti.com/ is working on a very common use of HPC tools by non-computer people. They are very focused on providing easy to use ultra low maintainance computational tools primarially for the bio-informatics community. One of the founders of the company worked an LANL on green-desitiny (or something like that) which was designed to be a low power low maintainance super computing resource at LANL. After all that, the short answer is yes performance is important but there is a lot of work and interest in making sure that this performance can actually be used by the people that are actually solving problems. Mark

    9. Re:Wake up, Bill by NardofDoom · · Score: 4, Funny
      You wanna connect to the internet via DSL? No problem.

      Come on now. You and I both know that's BS.

      --
      You have two hands and one brain, so always code twice as much as you think!
    10. Re:Wake up, Bill by Mr.+McGibby · · Score: 1

      combination of ease of programming for, ease of using, and of course running a data set on the machine.

      Which one of these has the biggest impact on overall performance? Certainly not "ease of using", which is what Bill says he's solved.

      --
      Mad Software: Rantings on Developing So
    11. Re:Wake up, Bill by P3NIS_CLEAVER · · Score: 0

      Is running gcc and eclipse on windows that much of a nightmare? Sure it is better on linux and with more utilities, but you act like writing a custom app on Windows is impossible.

      --
      Please sign petition to restore sanity to our banking system!!!

      http://financialpetition.org/
    12. Re:Wake up, Bill by k_greenwood1 · · Score: 1

      Can't say I disagree with you, but this should have some considerable repurcussions on what qualifies as "High Speed Cluster Administrator/Implementor" on a resume.

    13. Re:Wake up, Bill by drinkypoo · · Score: 1

      More scalable? I doubt it. They haven't succeeded in scaling up or down as well as Linux for any purpose, at least with NT (Windows 98 will give acceptable results on much less hardware than Linux) and I don't see why the clustering thing would be any different given that Linux has better network and disk throughput. More industry-accepted is possible, except that the "industry" in question is primarily scientific and not afraid of Unix.

      --
      "You're right," Fisheye says. "I should have set it on 'whip' or 'chop.'"
    14. Re:Wake up, Bill by schon · · Score: 1

      Is running gcc and eclipse on windows that much of a nightmare?

      Wouldn't know, I've never done it. Perhaps a better question would be "is running VisualBasic on Windows so easy?" And the answer would be "yes, that's exactly what I mean."

    15. Re:Wake up, Bill by XO · · Score: 1

      As if people use Linux on supercomputers?

      I had a guy arguing with me a couple days ago that Cray 1s ran Linux. flol.

      --
      "Champagne for my real friends - and real pain for my sham friends!" http://ericblade.postalboard.com/
    16. Re:Wake up, Bill by photon317 · · Score: 1


      Perhaps in academia, but don't forget that where some of us work, we're private users of supercomputer horsepower. And at least in my private area, it's still all about raw computer power per dollar (where dollars include hardware purchase, maintenance, administration, power, air conditioning... and yes, even programmer time, but see next sentence). However, our algorithms change infrequently, while bulk data to be processed and mined and whatnot comes in constantly. If we had to change our algorithms for every job, then perhaps we'd be back in the same boat as academia, but we don't.

      --
      11*43+456^2
    17. Re:Wake up, Bill by electroniceric · · Score: 1

      Good point. Not only is the supercomputer market a small one, as many others have pointed out, it's not a broadly lucrative space - how many scientific operations have spare cash lying around.

      However, below the weather prediction, protein folding and quantum chemistry type problems is a whole tier of computation on less-than-super computers. Nearly every scientific department at a university has a couple of these: 16-100 proc little gremlins that run people's models and analysis code. Not to mention automakers (computational fluid dynamics anyone) and financial modelers. And those gremlins ain't cheap either. MS is likely trying to establish their bonafides at the top of the market so they can sell to this next tier. If you run Windows for your domain and email services, your admins are more likely to be happy with the idea that you can buy a "proven" clustering solution that lets you use a GUI to add more boxen to a cluster. Given how far Linux has penetrated in the financial industry, this may be a losing gambit there, but it's my best guess of what the gambit is.

    18. Re:Wake up, Bill by rufty_tufty · · Score: 1

      "except that the "industry" in question is primarily scientific and not afraid of Unix"

      While i do agree with this I'm slightly worried about the PHBs who would be making the call to sign the chequebook.

      The PHB knows windows, they might not like this linux commie stuff.
      Besides MS took him out on a lovely free holiday, sorry, convention. Not that that will affect the decision...

      --
      "The weirdest thing about a mind, is that every answer that you find, is the basis of a brand new cliche" -
    19. Re:Wake up, Bill by vettemph · · Score: 1

      Agreed, why the hell would you want to give hackers a highspeed connection to your 'not fit for a network' workstation. :)

      --
      The government which is strong enough to protect you from everything is strong enough to take everything from you.
    20. Re:Wake up, Bill by Pharmboy · · Score: 1

      Please god tell me you are joking.

      "Windows" and "scale" are not used that often in the same sentence, although they are working on it. It can't even come close to Linux in scalability. Actually, only Unixes can come close.

      I will give MS the due they deserve, which is that they have a better desktop (once you untheme the crap out of it back to Windows 2000 mode) but in no way can they compare with serious computing yet. Workgroups and such, maybe, not my bag. But to scale is another story where Windows simply doesn't have hardly any experience yet. Maybe in 5 years or so, just like it took Linux several years to scale that large in a stable enough manner for enterprise.

      Also, whereas IBM, HP and RedHat and other companies and individuals have contributed considerable talent and time to get Linux to scale like it does, Microsoft is a single company, with a single focus. Lots of money to throw at it, but it is all coming from the same direction.

      One of the huge advantages of Linux is that it is developed by so many with different experiences. No one person or company can always know the best way to do something every time, so the fact that MS has the bucks doesn't guarantee they will get the best results, only that it will be the most expensive to develop.

      --
      Tequila: It's not just for breakfast anymore!
    21. Re:Wake up, Bill by Anonymous Coward · · Score: 0

      When you see [tt] in a post it means Troll Tuesday and it is a purposeful troll. You are wasting your breath.

    22. Re:Wake up, Bill by Anonymous Coward · · Score: 0
      yeah, there are plenty of custom apps in the HPC area, but there are also lots of off-the-shelf solutions that people could run on a cluster of windows boxes - Mathematica, Ansys (Structural analysys), Fluent (CFD), Gaussian (chemistry). These apps are usually distributed in a binary format and only support specific version of Linux - sometimes they can be a little difficult to get working correctly.

      I could easily see MS being mildly successfull with small clusters designed for researchers using off the shelf software I mentioned above.

    23. Re:Wake up, Bill by Anonymous Coward · · Score: 0

      HA HA HA! I wait to see it happening! first let them invent how to unlink open file~1 or allow two users to work simult. on the same computer! Supercomputing is for serious people, no space for toy OSes here, sorry.

    24. Re:Wake up, Bill by NitsujTPU · · Score: 1

      Let me be brief and blunt.

      1) What they are doing is available for other systems (ie Linux), in some places, though not as at fundamental a level in the OS. This is something that will have a major impact on performance.
      2) What they are doing is fast. By fast, I mean blazingly, outrageously fast. Application level benchmarks will be fast. They will be very very fast.
      3) By ease of use, they're not talking just about pretty graphical widgets, they're talking about implementation of distributed computing platforms. There are very few experts in the world on this topic, and even the experts disagree with each other on the finer points. Doing something with the correct algorithm in a distributed system will give you orders of magnitude better performance, just the same as in one that isn't distributed. Of course, most people didn't study distributed computing in college, and fewer in industry (assuming that not all programmers have degrees in computer science, let alone advanced degrees) have in industry.

      I can't express how important the ease of use factor is. What happens when a node goes down? You don't want someone tooling around with your cluster regularly. You want it to sit there and work. When you add stuff to it, you'd like for integration to be very hands off. Want a checkpointing algorithm? Want task migration?

      If you're talking about speed per dollar, then you have to account for development cost. We're not talking about office here. Nobody is developing your environmental simulation for free. If it's faster for your people to work, and they don't have to implement the lower level primitives required to support this functionality, you're going to save money... more than enough for a Windows license or 2, I promise.

    25. Re:Wake up, Bill by NitsujTPU · · Score: 1

      You can't just run your application on a supercomputer and get it to run faster. You have to have an application that is written for the supercomputer. If you save 10 weeks in compute time on the supercomputer, but it took you a year to develop the code, you didn't get a solution faster.

    26. Re:Wake up, Bill by Frumious+Wombat · · Score: 1

      Maybe. On the other hand, you have a group such as the AbInit team (solid-state quantum simulations) or NAMD (classical molecular dynamics) which writes the code, and then hundreds of different researchers actually use it. For the authors, a good programming environment speeds their time to use. For everyone else downstream, they want the absolute fastest system, as they are probably not doing serious work On the code, but rather With the code. They can put up with a somewhat user-surly environment as long as it runs as quickly as possible.

      Therefore, for most groups, "time to solution" is determined by raw cluster performance. This is the argument we used to have with C++ nuts from Comp Sci; they'd claim that we were missing out on all the great new software development methodologies by using Fortran, and we'd note that our Fortran code was at least 20% faster (and frequently much more) than theirs, thereby giving us an extra 2.4 CPU-months of runtime per year.

      This is the other place that Windows loses. Unless they have a posix-standard environment, nobody is going to develop code on their Windows desktop, and then run it on their local windows cluster, if they know that their job will later have to go through a wrenching porting process to a real supercomputer. Code written on OS-X/Linux can be (with generally minor adjustments) ported quickly to another larger Unix system (IBM SP3, Cray vector, etc). Anything that makes Windows-native calls, or uses some non-unixy shared memory model, is going to be a beast to port, and severely limit the future applicability of your program.

      I know the guys who run a big Windows cluster, and while I like them, what they're supplying is a pool of workstations. Not many parallel jobs get run, but a lot of serial jobs. It's easy for novices to run a few standalone jobs, but hard to migrate parallel code to, so most researchers don't. The real theory groups are all building their own Unix-based clusters, to avoid the Windows-based one. It was an interesting political decision to go with Windows, but it's not really displacing supercomputers or other clustering technologies.

      --
      the more accurate the calculations became, the more the concepts tended to vanish into thin air. R. S. Mulliken
    27. Re:Wake up, Bill by Salis · · Score: 1

      Unless it's an application that has already been parallelized and only needs a bigger / faster cluster to work on.

      Scientific and video editing/rendering HPC applications are frequently parallelized so the work has already been done!

      --
      Favorite /. tagline: "On the eighth day, God created FORTRAN." And it was good.
    28. Re:Wake up, Bill by NitsujTPU · · Score: 1

      Applications that were not already written for clusters will require development time. Along those lines, the vast majority of applications that will run on clusters haven't even been written yet. If the book were closed on computer science, then I would drop out and become a carnival geek... or maybe an astronaut.

      All tangents aside, seriously, software will be written in the future. If an API can help to bring them into existence faster, then that yields direct savings for the operator of the cluster, as well as allowing for their results to come more quickly. If that API also happens to offer a faster solution (which I believe this one does, in general), then there is no reason not to use it.

      Your assertion hinges on the idea that most of this software has been written already. I can tell you firmly, that unless the past century has produced more than half of the software that will ever run on clusters, this is certainly not true.

    29. Re:Wake up, Bill by Salis · · Score: 1


      HPC ppl don't care about software with new "features", but without new capabilities. Go to www.netlib.org and browse all of the HPC code that is already available and is readily reusable. It's quite a lot.

      And the APIs you speak of ... I wouldn't trust Microsoft to build scientific APIs for shit. They have this wonderful reputation for not caring about bugs ... and while you might not care about a bug that causes the occasional BSOD, a bug in a scientific API is more likely to produce wrong results. Yeah, that's huge. I won't even bother with MS.

      --
      Favorite /. tagline: "On the eighth day, God created FORTRAN." And it was good.
    30. Re:Wake up, Bill by NitsujTPU · · Score: 1

      I'm not talking about the scientific APIs. I'm talking about the clustering APIs.

      Beside that, I'm not even really talking about the specific implementation. Other implementations of virtual synchrony exist that could be exploited. Having kernel-level implementations however, will speed matters up.

      As for their clustering API. The underlying technology is designed to be resilient in the face of crashed and/or flawed machinery, and should be more than adequate to provide correct results, even in the face of non-byzantine failures at a number of nodes, depending on the implementation built over it.

      As for MS. I'm typing this to you from the Gentoo partition of my dual booted laptop.

      MS is not trying to sell you a scientific API. All of that will have to be implemented on top, and some people will be willing to do so. The theory center here has been running Windows clusters for some time now.

    31. Re:Wake up, Bill by NitsujTPU · · Score: 1

      Also, I'm not talking about "features", I'm talking about the ground level technology that people will need to use in order to develop HPC applications.

      As for HPC people... I'll ask my own HPC people how they feel about the matter.

    32. Re:Wake up, Bill by Salis · · Score: 1

      Heh, sure. If you ask them which OS they prefer for HPC, I will bet you $50 they say Linux (or some other Unix).

      --
      Favorite /. tagline: "On the eighth day, God created FORTRAN." And it was good.
    33. Re:Wake up, Bill by NitsujTPU · · Score: 1

      If you want to sound inflated about the matter and be a jerk... then fine.

      I'm giving a lecture on parallel distributed computing on Thursday to a class taught by one of the most influential voices in distributed computing. I'm trying not to drop names, because I'd rather not draw attention to my postings on Slashdot while I apply to PhD programs. Additionally, we've been running Windows clusters here for over a year now at our theory center, which is one of the best in the world. I am running parallel distributed computing experiments on my Linux laptop right now. Several of my best friends do research in parallel distributed computing. That's not research on a parallel distributed computing platform, they're actually researching the topics of scalability and robustness of scalable distributed platforms themselves.

      The research institutes here have produced both Linux and Windows software. I'm arguing that the techniques used to establish parallelism in Windows Cluster Edition are superior to shared memory techniques. Coming into this argument, you should realize that it is a rather excited issue in the systems community, about which you've established that you have knowledge only from the perspective of being a user.

      If you want to keep up at this, then you can, but you're being a jerk, and you honestly are arguing a weak point. I have friends on both sides of this debate. I already told you that I'm a Linux user. I'm telling you that this has the possibility to become rather popular, and if you want to tell me that I don't know what I'm talking about, then that's fine. I don't need to impress you. I'm doing just fine impressing the people that I need to impress.

    34. Re:Wake up, Bill by NitsujTPU · · Score: 1

      Jeez, I hate when I get all huffy. I'm trying not to say anything that I'll regret, being a person who believes in an overall principle of "being nice to others." Just don't take that tone with me. It's uncouth.

    35. Re:Wake up, Bill by Salis · · Score: 1

      You're right about me being just a 'user' of HPC systems. But what excites me most is not a new Windows cluster, but the type of massively parallel or Grid systems being developed, like the NSF TeraGrid. That is one cool system: multiple supercomputers at multiple Universities, sharing filespace through a gpfs-WAN, Grid-enabled with Globus, and each one with hundreds to thousands of cpus (mostly running Linux!).

      And new computing architectures like the Cell processors. That'll probably be the next level of scalable parallel (vector) computing, if their compilers can automatically vectorize/parallelize the code well enough.

      That's the type of parallel computing that excites me (as a user). What will MS/Windows do for me? How will they save me time or $? What will they offer that doesn't already exist or is not in the works? How good are their compilers? They have a long way to go before breaking the Top500 and I really don't think their clusters will be in demand in the meantime.

      It's cool that you're going for your PhD in parallel computing. It's a very interesting field. But, here's a tip for academia: Don't take any argument personally, especially if it's during the peer-review process. You'll save yourself a lot of unneeded grief. For example, going back through my conversation, there's no reason to call me a jerk. (And even if there is, you shouldn't ever get personal. It just looks unprofessional.) O well, I don't take it personally. ;)

      --
      Favorite /. tagline: "On the eighth day, God created FORTRAN." And it was good.
    36. Re:Wake up, Bill by NitsujTPU · · Score: 1

      Sorry about that. I generally don't get personal in matters. I didn't like the way that you argued "well, ask an expert." My gut reaction is, "ok, an expert taught me everything I know." I really really really shouldn't have overreacted. Still, I could point out that generally it's in poor taste to question a person's expertise, rather than allowing them to tip their own hand. I still should have pointed the matter out in a more tactful way.

      In short, I apologize.

      I worked for a few years and did well in industry, which is about the only reason that I can count on getting into a reasonable PhD program (undergrad wasn't a stellar experience for me). I just about had a heart attack when I posted with the word "jerk" in it. I guess that I separate slashdot conversations from professional ones, but, given the visibility, perhaps I shouldn't. I certainly don't want any hurt feelings. You seem like a reasonable person, and I'm always glad to befriend an academe.

      It's one of those things where I have a lot of conflict, because I have faith in the techniques used, because MS supported a lot of things that I've done in the past year, and because I'm also a supporter of Linux and the FSF.

      Still, I really shouldn't have lost my cool, and I apologize.

      The argument that I disliked was the "it's an OS vs OS thing." It really isn't. It's a technique vs technique thing. I thought about implementing the technology into the Linux kernel over the last summer, but I decided that thrusting myself into the center of a platform debate 6 months before PhD apps was a lousy idea.

      As for specific features of the OS... clue free. I think that the IPC substrate employed offers superior semantics, however, and that it offers a simple interface to program distributed code over. There are other implementations that have similar semantics (jgroups, spread). Time and $ savings is in the form of the easier interface. I also think that it will, in general, perform better on clusters with lower communication overheads. It may have problems scaling to a Top500 machine... but that will really be tied to the task that is being performed and the amount of overhead. MS surely can find a low communicationverhead distributed application that will allow it to hit the top mark if they like. That's really probably the key in winning that sort of thing, to be honest... though the program manager from BlueGene claimed earlier this year that he didn't see any limit on scale of that machinery (it kind of cheats by having separate processors just for communication, but what can you do, it's a nice approach).

      I would have to say, yes, grid computing and the like are very interesting efforts. They're based, however, on a different base technology. Grid computing is mostly virtual machine monitors. Distributed computing in the context that Windows Cluster will offer it is based on multicast distribution and synchrony models that deal with dissemination of information between machines. They probably SHOULD be related, but generally, they aren't (efforts have tried to tie them more tightly, but generally produce sort of niche application results... LINDA... Mosix).

      Interestingly, the PhD is in another area. Systems is just a sort of extremely avid hobby for me.

      Anyway, I really do feel a bit silly for flipping out over the matter. Generally I'm better about this. Still, I sincerely apologize.

    37. Re:Wake up, Bill by NitsujTPU · · Score: 1

      I should clarify that by communication overheads, I mean a sort of funny communication overhead that occurs in the specific technology used. Throughput is fine in general.

    38. Re:Wake up, Bill by Salis · · Score: 1

      There's really no need to apologize, but thank you.

      Yeah, most of my code is loosely coupled over multiple processors so a Grid-type environment with a high bandwidth, but (relatively) high latency connection is ok for me. I know a little about IBM's BlueGene system, but not much. They recently did some optimization of their dual core chips to allow the communicator proc to handle some computation when it's idle (I think this story was on /. with some cheesy heading.) They're starting to deploy BlueGene systems. The San Diego supercomputing facility recently received one. I guess it's still too early to see how well it performs on highly coupled parallel code (besides benchmarks). But we'll soon see!

      --
      Favorite /. tagline: "On the eighth day, God created FORTRAN." And it was good.
    39. Re:Wake up, Bill by idlake · · Score: 1

      So there is growing interest in the notion of "time to solution" as a combination of ease of programming for, ease of using, and of course running a data set on the machine.

      Ease of use on the admin side is achieved by hiring someone fulltime to maintain the machines; when you pay $1m+ for a supercomputer, you can afford to do that. So, even if Windows clusters or supercomputers were easier to administer (which they are not), that simply isn't an issue.

      For programming, most of the work goes into figuring out what you want to do and how to parallelize and distribute it; there is little that Microsoft can do to help with that, and, in any case, UNIX and Linux have the most advanced and mature tools in that area anyway.

    40. Re:Wake up, Bill by Bert64 · · Score: 1

      Not to mention that google are COMPETITORS.. You wouldn`t want your business depending on your competitors goodwill.

      --
      http://spamdecoy.net - free throwaway anonymous email - avoid spam!
    41. Re:Wake up, Bill by stienman · · Score: 1

      Supercomputers aren't about "Ease of use." They're about speed per dollar. When WCC can beat Linux on price/performance, then people will stand up and take notice. Not before.

      Here's a word problem for you:
      Suzy takes 12 hours to make the program that runs on the MS cluster in 12 hours
      Katy takes 24 hours to make the program that runs on the Linux cluster in 6 hours
      Both researchers have burdened costs of $70/hr for the university.

      Of course, the numbers are created only to illustrate that it is not a folly to pursue ease of use, even when the immediate goal seems to be performance. Other costs also determine the optimal solution, such as purchase price, maintenance, depreciation, etc.

      Even if they couldn't compete in performance (which they probably can) they can still gain the overall advantage by making supercomputing accessable to a wider range of research. IBM didn't think the world needed more than a few computers. In the 70s only engineers had computers - they were too hard for most people. Now a child can sit in front of a computer and use it - it's almost as popular as the TV and phone.

      Eventually each university will have a supercomputer. Then each college. Then each department at each university. While computers now are more powerful than supercomputers of yesterday, I still expect them to remain seperate logically - supercomputers will always be ahead of desktops for specialized applications. I doubt we'll have one in each home. But there is a lot of market left for the few supercomputer players, and MS wants some of the low hanging fruit.

      I have no doubt they will succeed to some profitable degree.

      -Adam

    42. Re:Wake up, Bill by CommieOverlord · · Score: 1

      What will MS/Windows do for me? How will they save me time or $?

      For many researchers, it will offer a familiar environment. It means you can test stuff on your desktop then just upload it to the the server and it works. That's BIG. The main selling point of Sun for HPC clusters is that the Solaris OS on the desktop is __exactly_ the same as the Solaris on the server.

      What will they offer that doesn't already exist or is not in the works?

      I've seen companies that had originally developed scientific/engineering tools for a Windows environment, and then later decide to parallelize them. Rather than start from scratch with a Unix port, they just modify the existing Windows app. If you're a researcher using that software you need a windows cluster.

      For using just using clusters as giant batch processing and that don't use parallel programming, they'll need a windows cluster if they're application was originally written for windows.

      I've seen researchers get bogged down ernormously in learning how to use a Unix cluster. Instead of doing research, they spent large chunks of time learning a new OS.

    43. Re:Wake up, Bill by Salis · · Score: 1

      For many researchers, it will offer a familiar environment. It means you can test stuff on your desktop then just upload it to the the server and it works.

      That's true. But that's also true for Linux. I do exactly what you say: I write code and compile on my x86 Linux box and then recompile and run on a large Itanium2 cluster.

      If you're a researcher using that software you need a windows cluster.

      What about the software requires a specific OS? For mathematical/scientific/engineering applications, most code is written in C/C++/Fortran. Very good compilers for Windows need to be available to run the code. While Intel's Compaq visual Fortran is available, no new versions are being developed. Will MS develop new compilers for the languages most used by mathematicians/scientists/engineers? People will not use .NET for scientific applications .....

      Linux has dominated the HPC community because a) it's free; b) free (or really good) compilers are available; c) people who develop the hardware for supercomputers (IBM, SGI, Cray, Sun, etc) tend to be Unix-centric (and going from Solaris/AIX/UNICOS to Linux is relatively easy); d) Linux has good IDEs for programmers (but the people who tend to develop HPC applications are not computer scientists, they're physicists, engineers, mathematicians, chemists, biologists, etc!); e) Linux is available for the desktop (vs. AIX / UNICOS); f) Linux has better security than Windows; g) Porting code from any Unix to Linux is relatively easy, (which is huge, because there is a LOT of HPC code out there. HPC people tend to reuse rather than redevelop). Most HPC code does not use too many OS-specific hooks because everyone wants their code to run on as many different computers as possible. Contrast this to Windows, where in order to get the most performance you almost have to use OS-specific hooks.

      Just a few reasons. Some more compelling than others.

      --
      Favorite /. tagline: "On the eighth day, God created FORTRAN." And it was good.
    44. Re:Wake up, Bill by CommieOverlord · · Score: 1

      That's true. But that's also true for Linux. I do exactly what you say: I write code and compile on my x86 Linux box

      Which doesn't help the people who use Windows on the desktop, does it?

      What about the software requires a specific OS?

      Sometimes it's stuff like the GUI not being sufficiently isolated from the computational code. Sometimes it is probably portable, but the time to port and test would take more time than any gains in performance. Or it's closed source and commercial

      Look, I use and like Linux and Linux clusters for HPC apps. I personally would never voluntarily touch a windows cluster. Windows clusters can't do anything a Linux can. However, a windows cluster may be, for some people some of the time, a better choice. Be it because of timing, training, or software restraints.

    45. Re:Wake up, Bill by Fished · · Score: 1

      Actually, many (most?) modern supercomputers run linux in a clustered environment. Cray's were a different story, of course, but the old Cray architecture is pretty much obsolete at this point.

      --
      "He who would learn astronomy, and other recondite arts, let him go elsewhere. " -- John Calvin, commenting on Genesis 1
  14. Easier to use for who? by Anonymous Coward · · Score: 0

    "Windows, even claiming it to be as powerful and easier to use than Linux" Easier to use for a M$-techwiz or for a Linux wiz?

    1. Re:Easier to use for who? by Anonymous Coward · · Score: 0

      I like the "as powerful as Linux part". Does this mean that (according to rumors of years past) Linux code has made itself into Windows for real?

  15. Now that CRAY is made by AMD by From+A+Far+Away+Land · · Score: 4, Funny

    The Opteron AMD processor is going into CRAY Supercomputers, so it only makes sense that Microsoft start making Windows for those AMD computers. What's next, a Beowulf cluster of Bill Gates?

    1. Re:Now that CRAY is made by AMD by The+Lynxpro · · Score: 2, Funny

      "What's next, a Beowulf cluster of Bill Gates?"

      I suspect that even with a Beowulf cluster of Bill Gates, the end results would still be described as "micro" and "soft".

      --
      "Right now, somewhere in this world, Scott Baio is plowing a woman he doesn't love," - Peter Griffin, *Family Guy*
  16. Hey that's great! by Anonymous Coward · · Score: 0

    Now you can tackle "cluster.log" corruption issue on failovers!

  17. Marketing by didit · · Score: 2, Interesting

    I've read many times here that having Linux in the top500 supercomputers list was not worth marketing because it is a niche. Now Microsoft is marketing a beta of what they dream might enter someday this list. Go figure ...

    1. Re:Marketing by einhverfr · · Score: 5, Insightful

      When I worked at Microsoft, I specifically argued *against* getting involved in HPC markets. It isn't really an interesting market. It isn't a big market. And it never will be. If anything it will get smaller rather than bigger. Yes, there are some applications that are not going away but these are not common. After all how many customers does Cray have? How many customers does Microsoft have? Ok, you have the answer to my question. Heck, the ISP and web presence provider markets are more important to Microsoft strategically than HPC.

      Indeed I cannot think of *any* reason why one would want Windows on an HPC cluster. Indeed, with Microsoft's reliance on COM and IPC stuff, I would be highly skeptical of using the Windows development environment in these cases. Yes, async I/O might be more mature on Windows, but I think that on the whole, Linux is a better choice.

      As for the ease of use factor. This is a product that is really only needed by a few highly technical people. Ease of use for beginners is not important here. Ease of use by experienced UNIX admins is. Sadly Windows fails here pretty badly. After all not everyone needs to build a Beowulf cluster with licensed Windows software in their basement and the intensive number crunching apps that such clusters are used for are the exception rather than the rule.

      Finally.....

      Why not take Windows Server 2003 Standard or even XP Pro (for fewer than 10 nodes), install SFU 3.5 and PVM and build your cluster that way? It seems that this would be better for the market than this new product which seems to be the worst of both worlds.

      This is just about saying "Anything Linux can do Windows can do better" rather than pursuing any reasonable business plan.

      --

      LedgerSMB: Open source Accounting/ERP
    2. Re:Marketing by Thud457 · · Score: 2, Funny

      I don't think 'botnets are commonly considered to be supercomputers.

      --

      the preceding comment is my own and in no way reflects the opinion of the Joint Chiefs of Staff

    3. Re:Marketing by vondo · · Score: 3, Insightful
      Indeed I cannot think of *any* reason why one would want Windows on an HPC cluster. Indeed, with Microsoft's reliance on COM and IPC stuff, I would be highly skeptical of using the Windows development environment in these cases. Yes, async I/O might be more mature on Windows, but I think that on the whole, Linux is a better choice.
      Why do Ford, Oldsmobile, Honda, you name it, get involved in Indy and Formula One racing:
      1. Prestige (branding)
      2. Research
      For whatever Microsoft spends improving Windows so that it can be used on today's supercomputers, the benefits they will reap for their server and workstation lines could easily repay that investment. I'm not at all sure they can make a go of it, but if they succeed it will help them a lot more than just selling some licenses for big iron.
    4. Re:Marketing by vondo · · Score: 1
      Sorry, my response was more to the first part of your post: Why would MS want to do this.
      When I worked at Microsoft, I specifically argued *against* getting involved in HPC markets. It isn't really an interesting market. It isn't a big market. And it never will be. If anything it will get smaller rather than bigger. Yes, there are some applications that are not going away but these are not common. After all how many customers does Cray have? How many customers does Microsoft have? Ok, you have the answer to my question.
    5. Re:Marketing by Anonymous Coward · · Score: 0
      For the life of me I can think of no advantage to running Windows for HPC but it is what we have come to expect. Microsoft are obviously desperate that Windows is not seen as lagging in any market segment but we've been here before. NT briefly took the workstation market and now everybody has switched back to *nix. In terms of procurement and platform strategy, NT was a very expensive mistake for everybody except Microsoft.
      Discounts, site licenses, lies; won't somebody please run Windows on a supercomuter? Pleeeeaaaaase.... we really want to leverage the PR....
    6. Re:Marketing by vettemph · · Score: 1

      It's not about making money in HPC. It's about shlong wagging in order to impress the CIOs of Corporate America. The money comes from retaining corporate desktops.

      --
      The government which is strong enough to protect you from everything is strong enough to take everything from you.
    7. Re:Marketing by patonw · · Score: 1

      It's probably about suits at MS seeing headlines every month "Linux takes 4 out of 5 top spots among world's fastest supercomputers" and asking each other "Hey! why aren't we doing that? Let's do that!"

    8. Re:Marketing by NitsujTPU · · Score: 1

      I don't think of it that way at all.

      Similar design primitives are used for high performance supercomputers as need to be used to run giant dot coms, replicated databases, and any number of other applications that do, or supposedly will, have a significant market. While I agree that OS licensing to these users will never match the desktop (by definition, there's no money to be made at a dot com if there are fewer users than dot coms), but introducing the technology here will give them the primitives to move into other markets.

    9. Re:Marketing by Anonymous Coward · · Score: 0

      They could probably boot Windows on number 5. (Looks like standard Intel Xeon servers from Dell with Infiniband interconnect cards installed.) Everything else "way up there" is either non-Intel architecture or weird enough that Windows probably won't boot. The top two are BlueGene systems which use a custom PPC based CPUs. Number 3 is IBM Power5 CPUs. Number 4 is Intel Itanium CPUs but I bet the architecture of these machines is SGI proprietary. Number 6 is AMD Opteron CPUs but, again, presumably Cray proprietary architecture for the rest of the machine. Number 7 is NEC vector processors. Eight is presumably IBM PPC970 blade servers. (Same CPU as the Apple G5 Xserve based systems at number 15 and 20.) Nine is another BlueGene system (custom PPC) and 10 is another Cray system. So in the entire top 10 there is maybe 1 system that could theoretically run this Windows HPC OS.

      That would make a heck of a press release. "Microsoft Windows Compute Clusterfuck Server 2003 can theoretically boot one of the worlds top 10 supercomputers*" (At the bottom of the page the * notes that this machine is actually running Linux.)

      The top two machines are essentially custom hardware. For very practical reasons (physical space, heat, power) when something comes along to overtake those machines I suspect it will probably also be custom hardware. In the near term maybe that means a faster BlueGene CPU module. In the longer term maybe Japan will build "Earth Simulator 2.0" and be the first to the 1 PF mark. That is certainly the next logical pissing match benchmark.

      At any rate the opportunity to be in the top 5 with commodity hardware has probably closed for a bit. Virginia Tech got lucky with System X...it was built while the supercomputing giants were all taking a short nap. (Not that there is anything wrong with the Apple hardware. If you don't need to be "way up there" the PPC970 provides one of the best GF/dollar ratios. When the dual core 970MP makes it into the Xserve or an IBM blade system we will probably see a few more 970 clusters.)

    10. Re:Marketing by Anonymous Coward · · Score: 0

      Lies!!! All lies!!!

      Oldsmobile will NEVER, and has NEVER been in Formula 1. Where did you get that crazy idea? Do you think they can actually afford it? Do you think they can build a quality car, or a race car for that matter?

      Well, if it did happen... Jordan and Minardi wouldn't always be at the very end of the starting grid.

    11. Re:Marketing by einhverfr · · Score: 2, Interesting

      For whatever Microsoft spends improving Windows so that it can be used on today's supercomputers, the benefits they will reap for their server and workstation lines could easily repay that investment.

      Well, I disagree with you here. To be a quality node in a supercomputer, I would think you would want a lightweight kernel with an efficient and simple programing environment. You want the ability to strip down everything you don't need and keep everything as simple as possible.

      This sort of design has absolutely nothing to do with any other market. It is sort of like the undersea aircraft carriers that Germany was looking at developing during WWII. You end up with something that is passable at either being a sub or an aircraft carrier and frankly lousy at being the other. Linux can succeed here because it is extremely modular and *is only a kernel.* Windows is a bit more than a kernel and is not so modular. So what improvements can be made?

      My point is that the improvements that could be made to Windows would be entirely useless in Microsoft's core markets. It doesn't even help them get licenses on big iron because the requirements are entirely different.

      The point is that Microsoft doesn't have to do any of this. As Windows becomes more stable and more scalable it is eating proprietary UNIX's lunch. Of course Linux is eating it faster, but that is another issue. Yet traditionally HPC has been the domain of a very small number of UNIX-like operating systems, such as UNICOS, UNICOS/mk, AIX, and a few others. Linux has been a more recent member of the family.

      --

      LedgerSMB: Open source Accounting/ERP
    12. Re:Marketing by Tough+Love · · Score: 1

      When I worked at Microsoft, I specifically argued *against* getting involved in HPC markets. It isn't really an interesting market. It isn't a big market. And it never will be. If anything it will get smaller rather than bigger.

      Well, somebody, I forget who, it was somebody who left Microsoft recently I think, came up with the theory that to beat Linux, Microsoft has to stamp Linux out in every single niche where it has any presence at all. That seems to be the reigning credo at Microsoft, hence this insane investment of resources into what looks to the rest of the world very much like a vanity product.

      It's stupid really. Microsoft cannot stamp out Linux any more than it can stamp out the internet. The main effect will be to give some of those ill gotten profits back to the income tax system, I suppose.

      --
      When all you have is a hammer, every problem starts to look like a thumb.
    13. Re:Marketing by einhverfr · · Score: 1


      Well, somebody, I forget who, it was somebody who left Microsoft recently I think, came up with the theory that to beat Linux, Microsoft has to stamp Linux out in every single niche where it has any presence at all. That seems to be the reigning credo at Microsoft, hence this insane investment of resources into what looks to the rest of the world very much like a vanity product.


      That was me. But I specifically exempted HPC from the mix because it is a small and highly specialized niche market that has very little to do with anything else. Compared to every other market, research done in this market has almost nothing to do with anything else.

      Secondly nobody in their right mind is going to run an entire HPC cluster on Windows for very long. Sorry, but Windows is just not well designed for this environment. Not that you can't run it there (you can with SFU, PVM, etc even without the Cluster Edition), but you it is not really optimized for it. Correcting for Windows' deficiencies in this area will be more complex and more problematic than any other market where Linux is commonplace. These markets include low-end web hosting companies, ISP's, and embedded markets (ok, this last one has more in common with MMP HPC than any others).

      --

      LedgerSMB: Open source Accounting/ERP
    14. Re:Marketing by Anonymous Coward · · Score: 0
      As someone with a background in embedded markets (including a WinCE - based media player in Taiwan that ended up going into production instead with a microkernel my team built; as well as vxworks, embedded linux, etc experience on systems ranging from medical imaging systems to cell phones) I'd say there's a lot more relevance in HPC for Windows than there is for embedded stuff.


      The range of embedded devices does not lend itself at all to Windows and the very limited range of hardware platforms it can handle. The systems I've worked on required everything from complier tweaks (where the instruction sets of cores were extended) to pretty big hacks to virtual memory subsystems (where dma-engine coprocessors would stream memory in anticipation of it being needed).


      Both are pretty easy on Linux - where you could complete the tasks in less time than it'd take to clear things through Microsoft Legal and get help recompiling their OS.


      Low end web hosting & ISPs is another area that Windows is an extremely poor fit for -- since the single driving factor there is cost. These guys specialize on finding $20 cheaper hardware than their competitors that makes the difference between profits and losses; and any OS that costs more than $1 or so is simply not even in the running. Yes, the margins really are that thin - since the whole industry works exactly the same way -- rent cages from a big carrier like sprint; fill them with wintel boxes; and install an OS. The only piece in that chain where there's any real price flexibility is the OS.


      At least in the supercomputing space, as the Microsoft guy the NYTimes quoted obsered, the benifits will trickle down into the mainstream.

    15. Re:Marketing by einhverfr · · Score: 1

      Well, I would agree that for most of the embedded market, Embedded Windows is an oxymoron. However, having supported Embedded XP and CE Platform Builder, I will tell you that the problems have nothing to do with limited hardware platforms. There are actually a large number of architectures that can be supported on these platforms.

      The actual problems are:
      1) Driver development is much more complex
      2) The Windows OS is not exactly lightweight, even though these toolkits are pretty modular. In particular the Win32 subsystem is usually what people are after when they choose these products for things like ATMs and the like.
      3) Licensing costs make this more problematic because they add too much per unit.

      So CE and Embedded XP are mostly reserved for handhelds and server appliances rather than real embedded systems.

      --

      LedgerSMB: Open source Accounting/ERP
  18. Is it just me by Anonymous Coward · · Score: 0

    Is it just me or are we going to see the biggest peice of bloatware ever produced?? Microsoft can't get it right on the desktop, how do you expect "WINDOWS" to work clustered.... Anyways I have a crisp $10 bill that says its going to be a flop in the market.

  19. *yawn* by Yahweh+Doesn't+Exist · · Score: 3, Insightful

    he's actually claiming it's *as powerful* as linux!? why should that interest anyone, unless it's also as free?

    1. Re:*yawn* by jcnnghm · · Score: 1

      Just a cluster of video cards. The only thing that has changed is the GUI.

      --
      You don't make the poor richer by making the rich poorer. - Winston Churchill
  20. Well, this makes sense... by Anita+Coney · · Score: 4, Funny

    ... as we'll undoubtedly need a cluster to actually run Vista.

    --
    If someone says he and his monkey have nothing to hide, they almost certainly do.
    1. Re:Well, this makes sense... by NardofDoom · · Score: 1

      Couple this with "Subscription Applications" and you have Microsoft single-handedly creating the Network Computer. That is, an OS that requires a network of computers to function.

      --
      You have two hands and one brain, so always code twice as much as you think!
    2. Re:Well, this makes sense... by master_p · · Score: 1

      No, it is about the upcoming Duke Nukem Forever.

  21. Piffle by the+eric+conspiracy · · Score: 2, Insightful

    even claiming it to be as powerful and easier to use than Linux.

    I find Linux ease of use to be perfectly acceptable, and since they are not claiming better performance, I don't see an advantage.

    1. Re:Piffle by surprise_audit · · Score: 1
      You really need to quote that in context to get the full impact:

      Gates, as always, has high hopes for this new version of Windows, even claiming it to be as powerful and easier to use than Linux.

      The way I read that, it implies that Gates himself believes previous versions of Windows have not been as powerful or as easy to use as Linux...

  22. NY Times Article (free reg. required) by __aaclcg7560 · · Score: 4, Interesting

    The NY Times has this article. The opening paragraphs were a bit more intriguing:

    "In January a group of Microsoft researchers set out to discover how much computing power they could buy for less than $4,000 at a standard online retailer. They found the answer at NewEgg.com, where they were able to purchase - for just $3,632 - 9.5 gigaflops of computing speed. That is the amount of computing power offered by a Cray Y-MP supercomputer in 1991 at a cost of $40 million."

    1. Re:NY Times Article (free reg. required) by steveit_is · · Score: 1

      Yeah, but that didn't include licensing costs :)
      Lower TCO my ass.

    2. Re:NY Times Article (free reg. required) by VJ42 · · Score: 1

      (free reg. required)

      Has th NY Times stopped asking people outside the US to register, becaause the last couple of times I've visited they havn't asked me, besideds, even when they used to, I used BugMeNot

      --
      If I have nothing to hide, you have no reason to search me
    3. Re:NY Times Article (free reg. required) by Anonymous Coward · · Score: 3, Interesting

      Yeah, its called an Apple G5 Xserve! (5GF rmax/CPU x 2 ~= 10GF for the $3000 cluster node) I'm sure the 970MP will make it into these at some point and that gets us 20GF/box (4 cores) Last time I checked you needed about 1000 of these boxes to get "noticed" on the Top 500 (see entry number 15 and 20.) If you can afford IBM Power5 boxes maybe you can take number 3.

      Somehow I think Windows Compute Cluster(fsck) Server 2003 does not run on the above hardware. Quick, there was just a November Top 500 list article posted yesterday...who was in the number 1 and 2 spot? Oh yeah, IBM BlueGene/L with 131072 custom PowerPC CPUs and a total rmax of like 280TF. Wow, that's only 30,000 *times* the CPU power of the MS $3500 off the shelf supercomputer. True, a 9GF Cray Y-MP may have cost $40mil 15 years ago. Today for $100mil you get 30,000 times the CPU power from BlueGene/L and can claim no 1.

      Can you imagine trying to put 15,000 "beige boxes" (I assume dual proc) purchased from NewEgg.com into a single room and have any kind of reliability? You think there might be a reason the "big" commodity CPU clusters top out at around 4000 boxes? (And that the hardware was not purchased from NewEgg.com...I mean WTF?)

    4. Re:NY Times Article (free reg. required) by fishybell · · Score: 2, Interesting
      Alright, maybe Microsoft wants to be on the Top500, but I don't. If they can offer a solution that, through clustering techniques, is more reliable, faster, and more expandable than their current solution, those of us who want to make the Top500000 (ie. what it takes to run a medium size enterprise database, a website that won't slashdotted to oblivion, etc.), $100,000 would probably get us there with 25 top-of-the-line, off-the-shelf boxes. That's the linux solution.

      Now take into account an extra $25,000 (wild-ass-guess) for licenses and you've got the windows solution. If it costs you $25,000 less a year to get a Microsoft trained IT team than a linux or Unix IT team, then this would seem like a decent trade off to the CFO and CTO; especially if they get to keep their same IT team with minimal training.

      From the NYT article: "Our focus is not on the very highest-end systems but on divisional and departmental computing systems," said Kyril Faenov, Microsoft's director of high-performance computing.

      --
      ><));>
    5. Re:NY Times Article (free reg. required) by Anonymous Coward · · Score: 2, Interesting

      So what we are talking about here is web servers and enterprise database systems? (Essentially high end "business" software.) Doesn't Microsoft already have software available for web server and database clusters? These are normally pseudo-realtime in nature. (In other words, when someone does a search on your website they expect a response now. Not a month from now.) If Microsoft doesn't, I'm sure Oracle will be glad to sell you a monster of a database cluster. I do realize database systems are used in other ways such as data-mining and that these types of applications are not realtime and are probably vaguely HPC. (Though I suspect you want IOs per seconds and not Mflops or Gflops or Tflops for this.)

      Normally I associate "supercomputing" or "HPC" with science and engineering. Finite element analysis, computational chemistry, that type of stuff. Maybe high end synthesis and verification for integrated circuits. Some people might try to argue that Hollywood style render farms (special effects) are a HPC application. Any computational process that takes a rack or possibly an entire room full of high end processors days, months, or years to complete might qualify as a HPC application of sorts. (Usually time to completion is proportional to the number of CPUs available if the computation can be parallelized else you probably own a fancy Cray or NEC vector machine instead of a parallel processing cluster.)

      I certainly got the impression that Microsoft is talking about this second variety of application because yes, this type of application does exist in many companies doing research and engineering. (Be that a biotech or pharmaceutical company, an automotive company, an energy company, a fables semiconductor company, etc.) Obviously you don't need to be on the top 500 if you are running these types of applications. (Though you might be, the COLSA system is being used for CFD / aerodynamics for military aircraft and missiles. Looks like number 12 is being used for biotech research.)

      Did I mention you get the OS for free with that Xserve cluster node? It's bring your own high performance interconnect and switches though. ;)

    6. Re:NY Times Article (free reg. required) by strider44 · · Score: 1

      I knew that Australia is part of the US now.

    7. Re:NY Times Article (free reg. required) by killjoe · · Score: 1

      What an interesting comment. I bet the open source crowd can use this and similar comments as advertisement. I can see the logo now "become an MCSE and make less money!". This could be a very powerful campaign to discourage people from becoming windows sysadmins.

      --
      evil is as evil does
    8. Re:NY Times Article (free reg. required) by Anonymous Coward · · Score: 1, Interesting

      $3,632 doesn't take into account the energy bill (to keep those systems up 24/7), the square meters of real estate you need to place those systems. Plus air conditioning is needed if you need a lot more than 9 gigaflops. Last but not the least, those systems may not compute 24/7 and they may break over time requiring expensive maintenance (software and hardware).

      Software solutions like http://www.cpushare.com/ havn't those hidden costs.

  23. Cause this makes sense! by The+Lost+Supertone · · Score: 2, Funny

    The question is, how much will your average cluster be spending on Norton as a result of this?

  24. Where could one find a drug by Hosiah · · Score: 1

    that could make one...soooooo deluuuuusional? Do we see Radio Flyer entering their wagons in the Indianapolis 500? KMart renting a space at Fashion Island mall? Weekly World News reporters showing up at the Nobel Literature Awards "just in case they win"?

  25. Zombie Cluster... by stay+bolt · · Score: 1

    I'll bet the spammers are already drooling at the potential of installing a rootkit on one of these babies.

    1. Re:Zombie Cluster... by Hosiah · · Score: 1

      *visions of "Attack of the 50 foot Spammerputer", complete with 1950's movie poster imagery* Spam pumped into my account at 1000's of teraflops per minute. Going to open my inbox, and the internal pressure makes the monitor explode in my face, accompanied by the AOL sound file. Thank you, my hair is completely white, now.

    2. Re:Zombie Cluster... by kryten_nl · · Score: 1

      I can see it now....

      Nodes $sys$000 to $sys$999 not found.
      [OK] [Cancel] [Browse]

      --
      For the perfect anti-Unix, write an OS that thinks it knows what you're doing better than you do and let it be wrong.
    3. Re:Zombie Cluster... by surprise_audit · · Score: 1

      What would be the point, unless it had an Internet connection to match the speed?? Having a zombie cluster wouldn't achieve anything if it spent large amount of time idle while waiting for the network to catch up.

  26. Will it have a GUI by CastrTroy · · Score: 1

    Will the supercomputer have a GUI? And will each node be required to run a GUI? That's one thing that I never liked about windows. Even your webserver and database servers have to waste resources running a GUI, even though it's perfectly possible to run such machines without a gui.

    --

    Anthropic principle: We see the universe the way it is because if it were different we would not be here to see it.
    1. Re:Will it have a GUI by middlemen · · Score: 0

      Forget the GUI. Who uses Windows for heavy computation anyway ?
      And even if people do (Excel spreadsheet VBA heavy computing by connecting to a heavily bloated MSAccess DB or an ORACLE DB), have those been ported or even tested on a multi-processor cluster to make use of the extra computing power ?
      Does VBA even have any capabilities without using the shady COM interfaces to do any networking or movement of data across computers ?
      Bill Gates is living a lie, waking up in the morning everyday and telling himself that Windows can do so much more than what Linux can do.

    2. Re:Will it have a GUI by sharpone · · Score: 1

      What would Windows be called if didn't display any "windows"?

    3. Re:Will it have a GUI by nick79au · · Score: 1

      ...DOS

  27. Linux by Anonymous Coward · · Score: 0

    Cool, but does it run Linux?

  28. Re:ummm, yeah, right.... by Foofoobar · · Score: 1

    Also make it use less than 32MB of RAM; right now XP sucks about 128MB RAM just to run Windows.

    --
    This is my sig. There are many like it but this one is mine.
  29. Gates' other endeavours by Anonymous+Custard · · Score: 1

    "even claiming it to be as powerful and easier to use than Linux."

    Gates is also authoring a new book called "Supercomputing For Dummies", for all those super-computer admins who are frightened by command prompts.

    1. Re:Gates' other endeavours by Farmer+Tim · · Score: 1

      ...for all those super-computer admins who are frightened by command prompts.

      And aren't already using Qmaster/X-Grid...

      --
      Blank until /. makes another boneheaded UI decision.
  30. This makes sense to me by Anonymous Coward · · Score: 0

    Have you seen the system requirements for Vista?

  31. Share nothing cluster? by Malc · · Score: 2, Informative

    Does this version continue to use share nothing and thus useful be mostly for high availablity? Or can resources now be shared concurrently between different nodes of the cluster and thus provide better performance?

  32. Yesterday's Future Tomorrow! by michael+path · · Score: 4, Insightful

    Nothing says "Product of the Future" like the beta version of a product named "Windows Compute Cluster Server 2003".

    Sure, they're doing it to maintain the "2003" branding of the flagship server. But why, less than two months before the end of 2005, are they not even trying to sound modern?

    1. Re:Yesterday's Future Tomorrow! by fbg111 · · Score: 1

      Heh, true, all they really need to do is just drop the 2003. It's not like the target audience can't see past the naming scheme anyway. (To them it could just as well be WCC NT 1996)

      --
      Flying is easy, just throw yourself at the ground and miss. -Douglas Adams
    2. Re:Yesterday's Future Tomorrow! by Anonymous Coward · · Score: 0

      Sure, they're doing it to maintain the "2003" branding of the flagship server. But why, less than two months before the end of 2005, are they not even trying to sound modern?

      MS knows that we all know that whatever they end up releasing is going to be years behind everyone else's software. They know they'll be the butt of endless jokes if they call it "Windows Compute Cluster 2006", so instead they call it "Windows Compute Cluster 2003", but it'll really be closer to "Windows Compute Cluster 1993".

    3. Re:Yesterday's Future Tomorrow! by plopez · · Score: 1

      Ever hear of planned obsolesence? Why upgrade? Beacuse your OS is sooooo 1998/2000/2003. No technical reason, you just have to be modern and up-to-date. Fashioanble, like everyone else. See also SQL Server 2000/2005.

      --
      putting the 'B' in LGBTQ+
    4. Re:Yesterday's Future Tomorrow! by jounihat · · Score: 1

      That's because for this young hip software company, tomorrow is today, and today is yesterday! You heard me.

    5. Re:Yesterday's Future Tomorrow! by Anonymous Coward · · Score: 0

      In 2008, Microsoft will stop supporting Windows 2003 because it will be considered 5 years old. Then, their five customers will upgrade to Windows Compute Cluster Server Vista for a nice ... Profit!

  33. It's no surprise... by Ted_Bell111 · · Score: 2

    Let us remember that much of Microsoft's revenue comes from the entire server/enterprise category, so it's no surprise Microsoft is going on the offensive and trying to compete with the power of Linux servers. Microsoft may act like its not scared, but clearly they are at least worried about the growing competition of Linux... especially if they always have to remind us how much "better (enter microsoft product here) is better than Linux."

    1. Re:It's no surprise... by MightyMartian · · Score: 3, Interesting

      Look, it's one thing to support small clusters. That's a reasonably profitable market place, and I can't imagine a modern OS that is marketed as a server solution not offering that feature. But what we're talking about here is supercomputer clusters, beasties used in nuclear weapons research, weather forecasting and other forms of computational-intensive work. This isn't exactly a huge market. In fact it's a downright small one, dominated by custom applications and by a few companies with a lot of years of expertise in high end computing. This seems more an example of the sort of megalomonia that runs in the bloodstream at Redmond. "Yeah, we gotta have a presence in the supercomputer market! How come no one's modelling black holes or doing long-range climatological forecasting on Windows 2003?" What do they think, that supercomputers are going to be running Exchange 2003? "Oh yeah, baby, look at how fast Excel comes up now!" I'm used to the idea that Microsoft is going to try to dominate huge sectors of the computer industry, but supercomputers? It's as if Gates and his toadies are losing their collective marbles.

      --
      The world's burning. Moped Jesus spotted on I50. Details at 11.
    2. Re:It's no surprise... by ChatHuant · · Score: 1

      Look, it's one thing to support small clusters. That's a reasonably profitable market place, and I can't imagine a modern OS that is marketed as a server solution not offering that feature. But what we're talking about here is supercomputer clusters, beasties used in nuclear weapons research, weather forecasting and other forms of computational-intensive work. This isn't exactly a huge market. In fact it's a downright small one, dominated by custom applications and by a few companies with a lot of years of expertise in high end computing.

      OTOH, a very similar point of view was expressed in the past about computers in general: computers are beasties, to be used indeed only in weather research, weather forecasting and computational-intensive work. Or, as the man said, "I can imagine a world-wide market for about five computers".

      Gates came with the "a computer on every desk" vision, and so we got to the current situation, where a current desktop-level Pentium or Athlon machine is many times more powerful than a military supercomputer of a few years ago. Given the continuous advance of technology, today's supercomputer will be tomorrow's desktop machine. So pushing the OS to deal with the (currently) high-level features of supercomputers, and making it easy to use seems like a reasonable strategy for future-proofing your software.

  34. I don't think so... by dslauson · · Score: 4, Insightful

    I have to say, I just don't think developers will go for it. First of all, if you're setting up a cluster of, say, 20 machines to run some MPI programs, you're going to be funneling some serious coin microsoft's way.

    Secondly, I, like many developers, have been running MPI programs on Linux clusters for some time now. What's my incentive to switch? All I've got is penalties, like having to buy software and stuff. MPI is already free, open source software. So now MS sticks it in their OS and sell it as a new platform?

    At least for me, this is too little, too late. I'll do what I've been doing, which is run my parallel code on Linux.

    1. Re:I don't think so... by nitroamos · · Score: 0

      the question is not what incentive do you have to switch... nobody will turn down computing power if it's cheap enough. if microsoft promotes their machines by seeding a couple highly discounted machines at well known research centers, then they have a way in. this is regardless of the quality of the offering. well, the worse it is, the more microsoft would have to discount it at first.

      e.g. a grad student's advisor would say -- look, we have all this new computing power that nobody else is using! go port your code!

    2. Re:I don't think so... by Salis · · Score: 1

      Actually, this is what Microsoft tried to do at my University. They donated a Unisys 32 proc Itanium2 cluster with Windoze loaded. After 6 months of non-use, our techies loaded Linux on 16 procs and people started to use the Linux portion. Still, no one uses the Windoze portion. O well.

      In addition, MS will not only have to donate the computers, but also provide _top-notch_ C/Fortran compilers for Windoze. Yeah, I'll be holding my breath....Intel has already beat them to punch by years.

      --
      Favorite /. tagline: "On the eighth day, God created FORTRAN." And it was good.
    3. Re:I don't think so... by Anonymous Coward · · Score: 0

      Well obviously developer's won't go for it. Nor will sysadmins. But I've never seen a developer or a sysadmin cough up millions for a supercomputer. That's a management decision, and as far as the management is concerned 'Microsoft' sounds better than 'Linux'.

  35. Right.... by Anonymous Coward · · Score: 2, Funny

    And if my grandmother had wheels she would be a trolley.

  36. until by akhomerun · · Score: 3, Insightful

    bottom line is, until microsoft can build this OS to be HUGELY FASTER than linux, there's no reason to pay extra for something that doesn't have any speed advantages.

    i've never heard of the supercomputing crowd complaining about ease of use, they are looking for more calculations for less money, and for that linux/unix is probably still the best choice. there's no reason to pay thousands for an OS that doesn't increase your performance any further than an OS that costs $0

    1. Re:until by rufty_tufty · · Score: 1

      Have you ever had a PHB that choose the wrong solution regardless of advice from his employees because of the better salesman/familiarity/FUD?

      I have and they would choose windows over linux if they could.

      --
      "The weirdest thing about a mind, is that every answer that you find, is the basis of a brand new cliche" -
    2. Re:until by idlake · · Score: 1

      bottom line is, until microsoft can build this OS to be HUGELY FASTER than linux, there's no reason to pay extra for something that doesn't have any speed advantages.

      They can't. Linux is so widely used and so heavily tuned that it's pretty close to hardware speeds in most areas that matter to current software.

      There are some obscure features where Linux doesn't perform well, but as soon as they stop being obscure (i.e., as soon as people start using those features widely), their performance will be improved.

      i've never heard of the supercomputing crowd complaining about ease of use,

      Among other things, that's because most supercomputers come with a naturally intelligent, voice operated management interface.

  37. Re:ummm, yeah, right.... by sundancekid503 · · Score: 1

    Puppy Linux makes a joke out of WinXP because it's smaller? If thats the case my DOS boot-disk makes a joke out of the Puppy Linux Distro.

  38. Re:ummm, yeah, right.... by Codename_V · · Score: 1

    With large high performance clusters, and even small ones if you're smart about it, it doesn't matter at all how big or small your installation media is. It's about automating the installations of your compute nodes as much as possible. For example with Rocks, you just pxe boot or pop in the install cd and the cluster nodes intall themselves completely hands off. Other solutions don't even require the compute nodes to have a hard drive (Oscar springs to mind). I'd fully expect the Microsoft solution to be similar. They'd be absolute morons to try and market something that wasn't. Of course the thing I wonder about is how the pricing will work, and how they expect to make it work if they charge you for each machine in your cluster. I mean, why would you pay one hundred dollars a node for hundred nodes when you could install Linux on each machine for free?

    --
    Free will is just an illusion
  39. Two quotes by dtfinch · · Score: 1

    $1 million for this semi-supercomputer running the tried and tested server operating system Linux, under an open source license. Or $2 million for the same supercomputer running a server variant of the popular desktop operating system Windows, same performance, assuming advertisements don't lie, and under a license which limits the possibility of tweaking. Which will the client choose?

    1. Re:Two quotes by CCFreak2K · · Score: 1

      Not to mention the fact that the Windows product is in beta (although you can arugue that some Linux stuff might be, you can roll back to older but stable software).

      --
      "Beware of he who would deny you access to information, for in his heart he dreams himself your master."
    2. Re:Two quotes by megrims · · Score: 1

      $1000 for this desktop running the tried and tested operating system Linux, under an open source license. Or $1300 for the same computer running a flashy variant of the popular desktop operating system Windows which performs slightly to a good deal less efficiently, assuming semi-popular opinion doesn't lie, and under a license which limits the possibility of tweaking.
      Which will the client choose?

  40. Re:Claiming? by VJ42 · · Score: 2, Interesting

    Really? I've only just started switching from Windows to Linux, I have a dual boot Windows XP Pro\Linux Ubuntu machine, the one and only problem I've had is connecting to the internet via my wireless card.
    My brother had exactly the same problem on his Windows PC, i.e we can both see the WAN, but can't connect to the internet via the router. He solved his problem before I started installing Ubuntu, so the two arn't related, but Windows is far from perfect, and Linux is far from the disaster you paint it as.

    --
    If I have nothing to hide, you have no reason to search me
  41. Super Windows for Computers by NoSuchGuy · · Score: 1

    Hi Bill,

    How about you develop a "Super Windows"(TM) for normal PCs instead?

    Why a Windows for Supercomputer? Because they make superior bot nets?

    These are just some questions

    --
    Grundgesetz * 23. Mai 1949 - 30. November 2007 - http://www.vorratsdatenspeicherung.de/
  42. The first hit of heroin's always free... by turgid · · Score: 2, Insightful

    I'm sure Microsoft will do what it always does, and cheat.

    Bill and Steve will see to it that a high-profile research centre (e.g. a university) will get a free supercomputer with a free Supercomputer Edition(TM) of Windows(TM) to play with and there will be much fanfare and positive publicity in the press.

    Just like when SGI and intel gave NASA a free 10240-processor Altix (made of itanics).

    1. Re:The first hit of heroin's always free... by Anonymous Coward · · Score: 0

      The Altix was actually something impressive. This.... not so much.

    2. Re:The first hit of heroin's always free... by triso · · Score: 1
      Bill and Steve will see to it that a high-profile research centre (e.g. a university) will get a free supercomputer with a free Supercomputer Edition(TM) of Windows(TM) to play with and there will be much fanfare and positive publicity in the press.
      Quite right! From TFA: :"To support its move Microsoft announced "a multiyear, multimillion-dollar investment in the academic community". It is bankrolling the establishment of ten institutes for High-Performance Computing with universities including Stuttgart(Germany), Southampton(UK) and Nizhini Novgorod State University(Russia)."
    3. Re:The first hit of heroin's always free... by zerocool^ · · Score: 1


      Fix.

      Heroin Fix. Hit of acid.

      --
      sig?
    4. Re:The first hit of heroin's always free... by turgid · · Score: 1

      Surely it's only a "fix" the second time you take it? :-)

      /me ducks.

  43. It'll be a Complete Cluster, alright... by Anonymous Coward · · Score: 0

    Any ex-military folks know what a complete cluster is. You know, a complete cluster f**k?

    Who says there's no entertainment in advertising?

  44. The difference by everphilski · · Score: 1

    The difference is where supercomputing matters - areas of science and engineering - are dominated mostly by Linux but also have contendors in Windows.
     
    (Engineering is highlighted because that is my area of expertise - company I work for does high fidelity simulation in both Linux and Windows... but few if any companies in the engineering world work under OS X)
     
    -everphilski-

    1. Re:The difference by FLAGGR · · Score: 2, Insightful

      Yeah, rendering, video editing and the other things Apple's xserve's don't need supercomputers.

      Science and engineering are *not* the only places where supercomputers matter. You're just bias. Just watch the making of documentary for epIII of star wars, and look at all the shiny G5's hooked up to xserve's with awsome apple cinema display's. Science and engineering may be tilted heavily to windows and linux, but movie editing, 3d rendering etc are even more so geared to osx (well, not as much 3d rendering as movie editing but still)

    2. Re:The difference by Anonymous Coward · · Score: 0

      Macs have quite a following in Biology. My university is one of the world leaders in genetics research and we have a couple Xserve clusters floating around campus to support that research.

    3. Re:The difference by RzUpAnmsCwrds · · Score: 4, Informative

      Just watch the making of documentary for epIII of star wars, and look at all the shiny G5's hooked up to xserve's with awsome apple cinema display's.

      Better yet, watch the end credits - look for the huge AMD logo. Episode III was rendered on Opterons, not XServes.

    4. Re:The difference by NitsujTPU · · Score: 1

      Actually... that was marketing. Macs were too expensive for home, and too expensive for business, they managed to take hold in the educational and artistic sectors. They sell style with your ipod, which is why they are able to DRM the whole iTunes bit so it only works with iPods, and still sell the combo like hotcakes. Want proof? Look at Napster. I buy an mp3 from Napster that will work on any device that I want, and it costs me a buck. If I only want to listen at my computer, then I have a subscription service. If I don't want to pay $400 for a Napster brand NapsterPod, well, I don't have to, I can get a number of 3rd party players that work with their upgraded subscription service, and, as I said before, the mp3's I purchase have no DRM on them.

      So, why is it that iTunes making commercials on TV and paying the bills, and Napster is just kind of chugging along?

    5. Re:The difference by CaseOfThaMondays · · Score: 1

      or even better watch the latest movie from pixar, rendered on intel machines.

      --
      thats pretty much my best post ever. I spent like 3 hours typing it.
    6. Re:The difference by Anonymous Coward · · Score: 0
      I can speak for the ~7000 computers we have on our engineering campus. A majority of them are Windows ( no surprise here), but over the past few years Mac OS X has increased dramaticaly. In fact Mac OS X equals all the Linux and Unix boxes put together 1188 Mac - (948 Linux - (530 -53)dual boot + 545 Unixes ). Most of us are replacing our Sun, HP, SGI machines with Mac OS X, Linux and some BSD boxes.

      The numbers: 1188 Macs, 3453 Windows, 948 Linux *, 545 Unixes
      * 530 of the 948 Linux machines are dual boot with Windows. Over 90% of users choose to boot with Windows versus Linux
      ** 600 machines are unknown exotic OSes floating on campus. BSDs, QNX, student brewed OS, honeypots...etc.
      ***1000 laptops unregistered with depts floating around with students and researchers which only increases the number on Windows and Macs.

      You want high fidelity? I got your fidelity right here World's Highest-Resolution Display Wall and 200 Mega Pixel HiPerWall

    7. Re:The difference by NardofDoom · · Score: 1
      Bioinformatics is where most of the new Mac jobs are being created, and that's one of the target markets for Apple's XServe Cluster, which is specifically designed for use in a clustered environment. That and the ultra-high-end professional video editing and 3d animation markets.

      \Imagines 84 2.3Ghz processors in one rack
      \\faints.

      --
      You have two hands and one brain, so always code twice as much as you think!
  45. Re:ummm, yeah, right.... by jcr · · Score: 2, Funny

    The Puppy Linux distro makes an absolute joke of Windows XP....

    Since when is another OS required for Windows to be joke?

    -jcr

    --
    The only title of honor that a tyrant can grant is "Enemy of the State."
  46. Re:Claiming? by LnxAddct · · Score: 1

    I've had more success with usb and audio under linux than on windows. Sounds to me like you don't know what you are talking about. Use a distro designed for the desktop like Fedora, Ubuntu, or Mandriva.
    Regards,
    Steve

  47. Bill has a point. by everphilski · · Score: 1

    A fast computer that isnt usable is useless. Like a fast car that isn't street legal.

    If Microsoft can make inroads into newer supercomputing arenas with newer people who don't want to learn Linux, etc... they may have a market. I say it half sarcastically because I agree with you, yet I can see where Microsoft is taking this.

    -everphilski-

    1. Re:Bill has a point. by indigoid · · Score: 1

      tripe. fast** cars are pretty useful on racetracks...

      ** not necessarily meaning "powerful" - the lovely Lotus Elise / Exige are superb examples, as is any Lotus Seven (or Seven clone)

      --
      P-plate adventurer
    2. Re:Bill has a point. by jsveiga · · Score: 2, Funny

      Specially if it comes bundled with Microsoft Visual Basic for Supercomputing - Science Edition, allowing any computer-illiterate scientist to easily put together simulations without needing to care about multiple-processor optimizations. It will have ready-to-use MSAtomicNuclei.ocx, MSAtmosphere.dll, etc.

      You only need to go: Create a form. Drag the AtomView object to the form.
      Dim MyAtom(1E100)
      For i = 1 to 1E100
        My Atom(i) = MSAtom.new(x,y)
      Next
      Atom(0).Nucleus.Split

      Who wants to learn C, and all those threads things??

    3. Re:Bill has a point. by jsveiga · · Score: 1

      s/My Atom|^Atom/MyAtom/g
      Well, it's a computer-illiterate scientist's program...

    4. Re:Bill has a point. by Paul+the+Bold · · Score: 1

      You want to run an interpreted language on a supercomputer? Why not hitch a pony up to a Porsche?

    5. Re:Bill has a point. by FLAGGR · · Score: 1

      You want to run a Microsoft application on a supercomputer? Why not hitch a tub of nitroglycerine to a billion dollar check and throw it out the window (into a fire pit, incase the impact doesn't set it off)

    6. Re:Bill has a point. by Pharmboy · · Score: 1

      Dammit, and I had JUST used my last mod point. Good show.

      --
      Tequila: It's not just for breakfast anymore!
    7. Re:Bill has a point. by FLAGGR · · Score: 2, Insightful

      Uh, okay. You do realize super computers are EXTREMLY costly? There will always be enough experts out there to design an app for you. A scientist isn't likely going to write his own app for a super computer, he would hire someone that could make it work the best. A graphics designer isn't going to build their own distributed rendering software, they're going to get it made from someone else. Your fast car analogy is crap, because it's implying that Linux isn't usable. Just because you can't install it, doesn't mean someone who makes a living doing it can't. In fact, Linux is sooo much easier for something like this over windows it's insane. Only if you forget the fact it's impossible with Windows as it is. MS would be VERY hard pressed to beat Linux is the supercomputing market. Since it is open source, it's very flexible. You can get an entire system, with command line tools, on a floppy disk at around 600kb. Or you canget a run-off-the-cd or even DVD distro, like knoppix. All that and once you've figured thing's out, it's very easy! (Just read the online Linux-From-Scratch tutorials, you can literally take the source for apps, compile it, bootstrap and have your own custom OS as easy as pie)

      There aren't going to be newer people who want to build supercomputers but refuse to learn Linux, because there are alrady so many that are linux proficent. If a CEO decides to make a supercomputer, he doesn't make it, he hire's someone. There are plenty of someones who know Linux. It'll be faster in Linux, and believe it or not programming an application in linux and dealing with open source API's is much easier than windows (what the hell is error: 0x432d423a supposed to mean?) so you get faster code and faster development. What advantage does windows offer?

    8. Re:Bill has a point. by Bingo+Foo · · Score: 0

      Wow! You sure do have confidence in the public's interest toward your evaluation of esoteric automotive specimens to which you will never have access.

      --
      taken! (by Davidleeroth) Thanks Bingo Foo!
    9. Re:Bill has a point. by jsveiga · · Score: 2, Funny

      > You want to run an interpreted language on a supercomputer? Why not hitch a pony up to a Porsche?

      Sorry, you're right. Why would anyone run poorly designed, bloated, unsuited software that could make even the fastest hardware seem slow?

    10. Re:Bill has a point. by ranton · · Score: 1

      He isnt saying that Microsoft is necessarily better than Linux for supercomputers. He is only saying is that there could very well be a market for them as well as linux servers. If that is the case then it makes sense why Microsoft would want to sell a few more peices of software.

      --
      -- All that is necessary for the triumph of evil is that good men do nothing. -- Edmund Burke
    11. Re:Bill has a point. by indigoid · · Score: 1

      tripe, again.

      http://www.prb.com.au/ - a superbly built Seven clone to be had for about 35k aussie. This is well within the "toy budget" of many people in the IT industry, mine included.

      --
      P-plate adventurer
    12. Re:Bill has a point. by indigoid · · Score: 1

      http://www.prbaustralia.com.au/ - preview mode is handy :/

      --
      P-plate adventurer
    13. Re:Bill has a point. by Knetzar · · Score: 1

      Actually, if you've seen some of the stuff MathWorks has, creating complex math simulations can be as simple as dragging and dropping equations/operations and connecting them.

    14. Re:Bill has a point. by Bert64 · · Score: 1

      Supercomputing is a specialist market...
      The Formula-1 teams have fast cars which aren`t street legal, Formula-1 is a specialist market too.

      --
      http://spamdecoy.net - free throwaway anonymous email - avoid spam!
    15. Re:Bill has a point. by CommieOverlord · · Score: 1

      A scientist isn't likely going to write his own app for a super computer

      Yes, actually they. Frequently

      You do realize super computers are EXTREMLY costly?

      A modern cluster is cheap.

      he would hire someone that could make it work the best

      Har, on a university professor's grant money?

  48. Metrics: BSPS by behindthewall · · Score: 4, Funny

    Blue Screens Per Second

  49. Re:ummm, yeah, right.... by VJ42 · · Score: 1

    Indeed, Puppy Linux gave my old 128 mb USB pen a newc lease of life, I now have an entirely portable operating system and It means I can still get work done if my uni network goes down (as is frequently the case).

    I'm sure running your own OS would be a breach of our 'terms of use' for using the uni computers, but I don't think they thought about it when writing them ;)

    --
    If I have nothing to hide, you have no reason to search me
  50. If you want easy of use pick Apple by Yes+BlueBerries · · Score: 1

    A year ago Apple computers made number 7 it the top list of supercomputers http://www.top500.org/lists/plists.php?Y=2004&M=11 the story of them losing ground when they were higher is http://www.macnn.com/articles/04/11/08/vt.falls.to .7th/


    If I wanted to go with the easiest computer operating system to use, that is heavy on graphical user interface, I would pick a Mac. It said in the ranking list: "System X 1100 Dual 2.3 GHz Apple XServe/Mellanox Infiniband 4X/Cisco GigE / 2200 Self-made"


    Got to go.
    1. Re:If you want easy of use pick Apple by P3NIS_CLEAVER · · Score: 0

      all your easy belong to us

      --
      Please sign petition to restore sanity to our banking system!!!

      http://financialpetition.org/
    2. Re:If you want easy of use pick Apple by Locke03 · · Score: 1

      No, no....
      All your easy are belong to us.
      Sheesh, get your engrish right.....

      --
      I don't care what youre doing so much as the idiotic way you're doing it.
  51. MS Cluster server.... by Anonymous Coward · · Score: 0

    ...when one BSOD is not enough...
    ...why settle for one BSOD when you can get a whole cluster of them...
    oh, btw... does MS magnificent cluster server still have swithover times measured in seconds ?
    --
    Disclaimer: I don't like Microsoft and I hate both Bill Gates and Steve Ballmer and I am not ashamed of it...
    thx. for your time.

  52. Benefits of the GUI by everphilski · · Score: 1

    If you have a windows admin worth his salt, the overhead of the GUI is negligible, and the benefit of the GUI (being able to remote desktop in or switch the KVM to use good graphical tools that help you solve problems quicker. Like it or not there are good graphical tools when used with good console tools that can help you solve problems quickly) is irreplaceable.

    -everphilski-

    1. Re:Benefits of the GUI by schon · · Score: 1

      If your admin is truly worth his salt, he won't *NEED* a GUI to work remotely - a shell is more than sufficient.

    2. Re:Benefits of the GUI by russianspy · · Score: 0

      With x-forwarding you can still use graphical programs without running a local X-server. As long as the libraries are installed. It's simple - connect to one of the computers with ssh, run the program and the window (if any) appears on your local computer. I mean if you're administering any kind of a serious cluster, your interconnect is going to be fast, so that's not a problem.

      At most what you're loosing with this system is a few megabytes of hard drive space. It takes no resources unless it is in active use.

    3. Re:Benefits of the GUI by Anonymous Coward · · Score: 0

      I'd support this comment for single/few machines. The trouble is it doesn't scale. If I have 75 compute servers to manage, I do not want to have to use a local GUI to administer them. I want a centralized management interface to ensure the same configuration is on each system. The scriptability of *NIX helps out in this instance. There are some good tools for doing the same thing under windows (and further some that are multi-platform), but none of them use the local GUI to get the job done.

    4. Re:Benefits of the GUI by triso · · Score: 1
      If you have a windows admin worth his salt, the overhead of the GUI is negligible, and the benefit of the GUI (being able to remote desktop in or switch the KVM to use good graphical tools that help you solve problems quicker. Like it or not there are good graphical tools when used with good console tools that can help you solve problems quickly) is irreplaceable.
      Egad! Bite your tongue; you must be new here.
  53. So Is Sony by Anonymous Coward · · Score: 0

    So Is Sony

  54. Double the Clusters by kg4gyt · · Score: 1

    With this relase MS also stated that only half of the computing power will be available. Half will be needed simply to run the Operating System itslef, one processor will be needed for every processor managed.

  55. The Windows Compute Cluster 2003 Cycle by AFCArchvile · · Score: 4, Insightful

    1. PHB sees Microsoft adverspamming for Windows Computer Cluster 2003, and believes the drivel.
    2. PHB makes case to execs, gets capital for an 80-node WCC2K3 cluster for eleventy billion dollars, thanks to Licensing 7.
    3. Admins shake their heads in disdain, get the thing running, and walk away.
    4. Developers waste time and resources reinventing the wheel.
    5. Nodes start to get rooted because the admins didn't harden the system.
    6. Organized crime groups use nodes to DDoS websites in the name of extortion.
    7. ...
    8. Profit! (for Microsoft, at least).

    --
    "Ancillary does not mean you get to rule the world." --U.S. Circuit Judge Harry Edwards, speaking to the FCC's lawyer
    1. Re:The Windows Compute Cluster 2003 Cycle by Anonymous Coward · · Score: 1, Funny

      7 is obviously BSOD.

    2. Re:The Windows Compute Cluster 2003 Cycle by Anonymous Coward · · Score: 0

      Do you work for Gartner?

  56. Right and wrong. by Some+Random+Username · · Score: 1, Insightful

    The nonsense comments about the overhead of a GUI are retarded. Who cares if less than a fraction of 1% of my system is being used by something I don't care about? It simply doesn't matter.

    However, there is no upside to a GUI. It offers a way for developers to write software that is difficult and time consuming to administer, and requires a much better connection for remote administration than ssh does. I have never found a single graphical tool that helps me admin anything, they are always a pain.

    1. Re:Right and wrong. by Anonymous Coward · · Score: 0

      GUI makes sod all difference to processor but it still eats a decent chunk of RAM. And no swap file on a compute node. Sure it's only 64 meg or something but that's more chips to buy and more power/heat to worry about.

    2. Re:Right and wrong. by cbreaker · · Score: 2, Informative

      Ohh, I like a good GUI. Note the fact that I said "good." The very thing I love about UNIX is that you can, and probably should, do everything via a text/command interface. It's a lot faster and easier to admin large numbers of UNIX boxes then Windows, in my experience. Not to mention UNIX is quite a bit more flexible when it comes to networking and data sharing.

      However, there's some things that gain great benefit from a GUI. Any sort of large scale user management, especially with a directory services type system, tends to be easier on the admin when you can visualize the directory and user account placement within a GUI.

      I also like GUI based reporting - you get a much better sense of what's happening when you can see it in a chart or graph in front of you. Data visualization is good stuff. Being trapped in a GUI for administrative tasks isn't.

      Windows 2000 and 2003 server have both made a lot of progress when it comes to doing things on the command line but they still fall short of UNIX because of the fact that it's non-trivial to tie the various (and completely different syntaxed) tools together. UNIX shell scripts are easy- learning windows scripting is a lot less easy and more difficult to impliment on many machines.

      I'd rather start with a command-line based system and build a GUI on top of it, instead of the other way around. That way, you can always admin things 100% with ssh, customized to your organization.

      --
      - It's not the Macs I hate. It's Digg users. -
    3. Re:Right and wrong. by FLAGGR · · Score: 1

      Who cares if your cluster is wasting CPU time? Your cluster that could've cost millions to make? (Remember: this OS is *not* targeted at you, "Some Random Username (873177)", but big buisness and scientists, etc) GUI may not take up much cpu, but it takes up a nice chunk of RAM, and multiply that by however many nodes you need to administer, and wow, what a huge waste of money. Linux can run in command line, open the GUI when required, then close back down when it's finished. Try and close the GUI in windows and fall back into DOS. Oh wait, that doesn't make sense.

    4. Re:Right and wrong. by tskirvin · · Score: 1

      Let's say you've got a cluster of 100 tightly-coupled nodes running a parallel job, each with something eating 1% of their CPU time. What happens when one of those nodes goes into the 1%? The other nodes have to wait for that node to finish up...and they're not computing in the mean time. You haven't wasted 1% of one node, you've wasted 1% of *50* nodes.

      But wait! What about the other 49 nodes? They're going to have that 1% taken off of them, too. These times aren't going to be synchronized, much as you'd love for them to be. Instead, on average, you're going to lose 1-(1-0.01)^50 of your compute time waiting on that 1%. That's 40% of your cycles.

      What, 1% is unrealistic? Let's go to .1% then. With that and 50 nodes, you're wasting 5% of your cycles. This adds up.

      The moral of the story: don't run any background daemons on compute nodes. Or run less tightly coupled software, of course...

    5. Re:Right and wrong. by socrates09 · · Score: 1
      I'd rather start with a command-line based system and build a GUI on top of it, instead of the other way around.

      How about using that old DOS O/S and building a windowing system on top of that?

    6. Re:Right and wrong. by cbreaker · · Score: 1

      If DOS was a full modern multitasking OS that supported and utilized all new modern hardware, with a fairly uniform system layout and configuration - then sure, why not?

      --
      - It's not the Macs I hate. It's Digg users. -
  57. If you want easy of use pick *ANYTHING* but Apple by J.+Chrysostom · · Score: 5, Interesting
    We've got a VA Tech-style Apple cluster over the course of the last year here at our university, and let me tell you, that thing had serious problem. It works now, but it took over a year of fighting Apple's stupidity to make the thing function decently. And we're one of the first people to get an Apple cluster up and running for serious computation (unlike VA Tech who have been building it, running the LINPACK benchmark and taking it apart time and time again). Examples of Apple's stupidity:
    • Apple's NFS server only supports 64 simultaneous connection.
    • Apple attempts to explain that "real" supercomputers don't have networked file systems.
    • Disk I/O was slow as molasses. Apple suggests using the AppleTalk protocol to communicate to the file server to speed things up.
    • Disk I/O still slow as molasses. Grad student discovers the software defaulted to non-buffered disk output (flush to disk).
  58. Good gods... by Anonymous Coward · · Score: 0

    Pray they don't 'embrace' MPI and cripple it beyond rational sense and then stuff it down vendors' throats as a 'standard'...

  59. It's not the OS. by Beardo+the+Bearded · · Score: 1

    It's not the OS. Most of the computing power in a cluster goes to communication between the nodes in the cluster. That's true of any operating system you put on it.

    --

    ---
    ECHELON is a government program to find words like bomb, jihad, plutonium, assassinate, and anarchy.
  60. "Next week windows will be better" by MosesJones · · Score: 4, Insightful

    A standard line from Bill, "wait till you see Vista its better", "wait till you see compute its faster".

    Amazingly the press continue to take Microsoft at face value on annoucing their version as better when they don't release what they announce.

    So sure MS is better at supercomputers... I mean they have such a history in it, just look at the top 500 its just littered with MS boxes.

    This isn't Windows v Linux, this is MS Research v IBM Research. The people behind the CPU, Relational databases, reliable messaging and of course the huge amount of work on massively scalable computers. If MS had real ability they'd be working with the big processing boys from the goverment and weather prediction areas.

    --
    An Eye for an Eye will make the whole world blind - Gandhi
    1. Re:"Next week windows will be better" by Archtech · · Score: 1

      There isn't a single Windows box in the top 500 supercomputers at present. No doubt that could change, but would anyone want to buy a Windows supercomputer at a price that would allow Microsoft and the hardware builders a profit? Meanwhile, 372 of the top 500 - just about 75% - run Linux.

      As to "MS Research v IBM Research", it's not so clear-cut. Sure, IBM employees invented the relational database, reliable messaging, transaction processing and lots of other stuff. But have you checked out the people MS has been hiring? I wouldn't be surprised if it now has half of the experts who made those breakthroughs - although at their age they may not be doing much important new work.

      Then again, MS itself doesn't seem to take much from its research arm. The research guys keep coming up with brilliant new stuff, but most of it is incompatible with Windows, Office, etc. as they are today. Just as Dave Cutler designed Windows NT along the lines of VMS - modular, secure, reliable - only to have his work spoiled when Gates insisted on imposing the same old Windows 3 GUI on top of it. It's ironic: MS used to laugh at IBM for its cautious, fuddy-duddy image and the way it was tied to not breaking compatibility with its installed base. Now they have the same problem in spades. Heh.

      --
      I am sure that there are many other solipsists out there.
  61. Anthropic principle: We see the universe the by PigIronBob · · Score: 0, Offtopic

    Anthropic principle: We see the universe the way it is because if it were different we would not be here to see it. this is better known as the WEAK Anthropic principle

    --
    You never catch me alive
  62. am I the only one ... by ltwally · · Score: 1

    Am I the only one that misread that, at first?

    I could have sworn it actually read, "Microsoft Complete Cluster F**k Server 2003"

    And, if you think about it... my misinterpretation might actually be closer to the truth. ;)

    --



    /dev/random
  63. The Beowulf must be cycling in its cluster by Sheepdot · · Score: 1

    Man, I'd like to see a Microsoft-Windows-Compute-Server-2003 cluster of THOSE!

    Nah, nevermind. Just doesn't have the same sort of 'ring' to it.

  64. Microsoft's Supercomputer by FukYa · · Score: 1

    WOW, image how much spam mail those babies will be able to kick out!!!

  65. mayhem to the MAXX by Device666 · · Score: 1

    How cool it would be to see all these virusses take all the processor time.. Can you imagine how it is to see a DOS attack with a chain of 1000 clusters?

  66. First Thoughts by crawly · · Score: 1

    My first thoughts when I read this, where that Vista is going to really raise the bar for running requirements. Reality then asserted itself.

    --
    GCS/S d-x s+(+): a C++++$ UL+$ P+ L++$ !E--- W++@ N++>$ !o !K-- w++$ !O !M !V PS++>$ PE !Y PGP+ t+ 5++ X++ R tv b
  67. Re:ummm, yeah, right.... by Anonymous Coward · · Score: 0

    I'm sure running your own OS would be a breach of our 'terms of use' for using the uni computers, but I don't think they thought about it when writing them

    0MFG!!1!1! U R t3h t0t@l H@xX0r d00d! K33p r@g1ng agA1nsT teh M@cH!N3!!!!!!

  68. New Registry Parameter for WCC by grahamkg · · Score: 1

    ComputeCycleDelay, default value 4 microseconds.

    --
    Graham
    Linux - Fast Pane Relief
  69. Windows already runs some of the biggest supercomp by Anonymous Coward · · Score: 0
    The Windows Botnets already are some of the biggest supercomputers in the world.

    And they're so easy ten year old kids in pakistan can be certified to run these windows botnets.

    And more evidence they're easy to use - you don't even need to buy or own the machine to run a windows botnet.

    The above points are serious - Windows clusters fine (as proven by the botnets) and is easy enough (as proven by the ten year old script kiddies).

    The real reason they're nowhere in supercomputing is because often a supercomputer will have custom components (interconnects, etc) and without a large pool of people familiar with the source (like Linux has) you can't take advantage of your hardware.

    Also, no one in their right mind will pay > $0 for an OS on 1000+ compute nodes - where the *entire* responsibility of the OS is to stay out of the way. You certainly don't need DirectY or Clippy running on your compute node; and Windows really has nothing to offer beyond Linux/BSD/anything that might be relevant on a compute node..

  70. Maybe as good as Linux by msbsod · · Score: 1

    But at least Bill Gates does not claim that Windows Clusters are as good as VMS Clusters. That would be another 20 years of software development.

    Computer Center: We just installed Windows 2003 on our supercomputer cluster.
    Alt.2600: We just "bought" a supercomputer cluster for $30.

    http://www.theregister.co.uk/2005/11/10/password_h ashes/

  71. Man I read that wrong by bill_kress · · Score: 1

    It looked to me like "Microsoft competes in Supermarket"

    I really hoped to see an article about $25 copies of Windows at the check-out asile--right between the pulp gossip magazines and the kids-eye-level candy.

    Seemed like it might be their best marketing scheme ever--skip anyone who knows anything about computers and go straight for small chlidren and people who are impressed with Tabloids.

  72. So what happens when? by oztiks · · Score: 1

    The only problem i see with running windows pcs on supercomputers is the fact you need to reboot the damn thing anytime either a) it breaks b) it needs an upgrade software/hardware.

  73. Wow by Anonymous Coward · · Score: 0

    130 comments in and only one lame "Beowulf cluster" joke...

  74. Whenever I plan on building a supercomputer ... by bizitch · · Score: 1

    ... the key words I look for are "Windows Public Beta"

    --
    ---- "Logoff! That cookie shit makes me nervous!" - A. Soprano
    1. Re:Whenever I plan on building a supercomputer ... by Pop69 · · Score: 1

      I'm trying to work out how "Windows Public Beta" is different from a normal release for MS ?

  75. Correction by Zygamorph · · Score: 1

    Personally I think there is a big difference between entering and competing in the market.

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

      I'm sure they will be able to penetrate and saturate the market.

  76. Supercomputing 2005 conference by ChaosDiscord · · Score: 1

    Bill Gates is scheduled to give the keynote at the Supercomputing 2005 (or SC05) conference, which is going on right now.

    1. Re:Supercomputing 2005 conference by Blakey+Rat · · Score: 1

      Wasn't that at, like, 8:30 this morning? Something like 4 hours before the article was posted? A local radio station (107.7) sent an intern out to record Bill Gates' speech in the morning.

  77. Re:Claiming? by dana340 · · Score: 1

    Did anybody here paint linux as a bad thing?? That user may be excused.

    --
    "10001110101 - periodic table with a centerpiece of mind" -Clutch
  78. What do you call a Windows Cluster of 4096 Xeons? by flapdoddle · · Score: 1

    The world's biggest trojan attack platform!

  79. Easier to use by dajobi · · Score: 2, Funny

    This is great news, I've always had difficulty interfacing with my supercomputer. Now even Aunt Tilly will be able to run simulations of nuclear explosions!

  80. Defining moment? by reclusivemonkey · · Score: 1

    There is a defining moment in every great empire where it starts to decline. Seriously guys, this could be it. Xbox 360 and Media Centre are reasonably o.k. moves for Microsoft. Not great moves, Xbox was a huge loss for Microsoft, but this really does seem foolhardy. I liked the quote from the Register...

    Speaking without apparent irony Muglia said: "We've spent some time talking to Independent Software Vendors recently and the software community welcomes the arrival of a consistent environment to this area."

  81. Re:ummm, yeah, right.... by dana340 · · Score: 1

    hilarious.

    --
    "10001110101 - periodic table with a centerpiece of mind" -Clutch
  82. What if you're wrong? by eebra82 · · Score: 1, Insightful

    A lot of people seem to know better than the rest of us by saying that Microsoft is entering a business with too little, too late and God knows what.

    Remember when Microsoft went to war with Sony and Nintendo? A lot of people said it was a dumb move and that Microsoft would basically get kicked out by Sony. Look what happened, now they seem to put Sony at risk. Instead of being behind, they're AHEAD.

    I'm not a big MS fan but I hate it when people seem to neglect the fact that Microsoft is the biggest software company in the world because they've been doing the best job so far. Obviously, since they are in the lead. And please don't flame me on this one, I prefer Linux but I don't see my dad and my sister using Linux anytime soon.

    Point is, Microsoft ain't going into this market without a serious plan.

  83. clippy? by eck011219 · · Score: 2, Funny

    It's the army of bicycling paper clips, ready to do my bidding, that I look forward to.

    I, for one, welcome our ... oh, never mind.

    --
    It is pitch black. You are likely to be eaten by a grue.
  84. Clippy: "It looks like you want a super cluster." by khasim · · Score: 2, Interesting

    Seriously, how much "easier" will it be at the cluster level? With Linux, you don't even have to install the OS on the hard drive. It can run off of a CD.

    At this level, the extra wizards and such just don't matter.

  85. Re:ummm, yeah, right.... by Anonymous Coward · · Score: 0

    Perhaps I should clarify for your benifit; had they thought of it, I'm sure taht my uni would have put somthing in about using your own operating system in the terms of use of the network, however I don't think they ever though anyone would do such a thing, hence they forgot, or deemed it unnessary. So I'm not *tries to decypher AOL speek* "raging against the machine", nor am I breaking any rules, just using a usefll tool in the form of puppy linux to keep my uni computers usefull after having Win 2000 with a bunch of restrictions installed on them. The initial statemen was also slightly tounge in cheek, but you obviously failed to notice that. --posting AC because it's way off topic now.

  86. Defeats the purpose of a supercomputer by oztiks · · Score: 1

    The main issue about using windows for supercomputer technology is the flexability to custom build the kernel the chipset.

    No matter what ms does it wont compete against the computations of a custom built kernel like linux offers for that specific chipset.

    Other then that the macrokernel model with its blaitently lame design will never hold up against a monolithic design, or even a modular linux kernel.

  87. And what will they call their computing platform? by springbox · · Score: 1

    It better dang well not be the MicroComputer ..

  88. about time? by digitaldc · · Score: 2

    My initial reaction was, didn't they already have supercomputers? Why is it labled 2003 when we are 47 days from 2006?

    An excellent use for these MS supercomputers is to host xbox MMORPGs that enable 10 million players at a time reenacting the opening fight scene from Lord of the Rings: FOTR. Each player will run towards the center, clash, fight and most likely die within the first 30 seconds of gameplay.

    --
    He who knows best knows how little he knows. - Thomas Jefferson
  89. Oh wonderful. by WalterODimm · · Score: 1, Funny

    Now we get to have clippy bugging our scientists. I'm sure they'll love having to click "Just start predicting the weather" every time they launch Microsoft Weather 2003.

  90. Re:ummm, yeah, right.... by schon · · Score: 1

    Other solutions don't even require the compute nodes to have a hard drive (Oscar springs to mind). I'd fully expect the Microsoft solution to be similar.

    This brings to mind one of my favourite MS stories.

    Circa 1996, the company I worked for was transitioning a customer's LAN from Netware to Win95 (it wasn't our idea.) The clients were all diskless workstations, running from a single file server. The customer didn't want to buy new hardware, and MS told us that it was possible to do diskless booting with 95.

    After fighting with their network for over a week, MS decided that it wasn't possible, because the workstations had floppy drives. "That's not a diskless workstation! It's impossible for them to boot if they have a floppy drive!" (ignoring, of course, that the same hardware had been running flawlessly with Netware for the previous three years.)

    I can see them selling these cluser 'solution' licenses to some hapless university, then saying "Hey! Your nodes [don't] have hard drives - that's an unsupported configuration. Sorry, no refunds."

  91. You underestimate Bill Gates by MOBE2001 · · Score: 1

    Gates seems to have forgotten the audience in this case

    You underestimate the intelligence of the people at Microsoft. They want to create a new market targeted toward a different kind of people, those who would not normally use a supercomputer because of the technical difficulties. It's kind of like the way the GUI opened up computing to a lot of people who found a command line interface daunting, to say the least.

    1. Re:You underestimate Bill Gates by vsprintf · · Score: 4, Funny

      It's kind of like the way the GUI opened up computing to a lot of people who found a command line interface daunting, to say the least.

      Superclippy: It seems you're trying to factor large prime numbers. Would you like to engage the Microsoft Compute Cluster interface?

    2. Re:You underestimate Bill Gates by daremonai · · Score: 1
      ... It seems you're trying to factor large prime numbers...

      ObPedant: Uh, factoring primes is trivial: p = 1*p...

    3. Re:You underestimate Bill Gates by vsprintf · · Score: 1

      ObPedant: Uh, factoring primes is trivial: p = 1*p...

      a) Who was it that said that factoring large primes would be a computing breakthrough?
      b) Were you ever invited to a party? (N/y)
      c) If you answered "yes" to (b), were you ever invited to another party? (N/n)

      Thanks for stepping on the joke.

    4. Re:You underestimate Bill Gates by noamsml · · Score: 1

      d. does anyone aside from people trying to inflate their ego actually consider the number of parties to which a person is invited to be in anyway connected to one's merit? (N/n)

      If you want to make yorsef feel proud, superior, and righteous, at least do it by actually contributing to society.

    5. Re:You underestimate Bill Gates by Locke03 · · Score: 1

      Sense when does anything anyone say on slashdot contribute to society? I thought it was all about making ones self feel proud, superior, and righteous while belittling others....silly me...

      --
      I don't care what youre doing so much as the idiotic way you're doing it.
    6. Re:You underestimate Bill Gates by vsprintf · · Score: 1

      d. does anyone aside from people trying to inflate their ego actually consider the number of parties to which a person is invited to be in anyway connected to one's merit? (N/n)

      Yes. It's considered a measure of social interaction, and while not an indicator of one's intelligence or male genital organ length, it does tend to indicate one's education and sense of humor if nothing else.

      If you want to make yorsef[sic] feel proud, superior, and righteous, at least do it by actually contributing to society.

      Well, I don't know if my two kids could be considered a contribution to society, but they did manage to grow up without any of us killing each other or anyone else. Then there are my published articles, including the ones that were anthologized, as well as all the work I've done during my career. Pride, superiority, and righteousness have nothing to do with my comment. It was a joke.

    7. Re:You underestimate Bill Gates by Duhavid · · Score: 1

      Perhaps we have. There are a very large number of
      very intelligent people at Microsoft.

      But my mind boggles. What I have always thought that you
      would use something in the supercomputer class for is lots
      of number crunching, math intensive type computation.

      I am sure that ease of use has some place in all of this,
      but I cant help but imagine that it's far below performance
      and robustness and correctness.

      --
      emt 377 emt 4
    8. Re:You underestimate Bill Gates by Anonymous Coward · · Score: 0

      You could have just said nothing.

      Now you're all hot and bothered... sweaty, shining a little in some places... fogging up windows, you get my drift.

    9. Re:You underestimate Bill Gates by Anonymous Coward · · Score: 0

      Heh, somehow I doubt these guys find "a command line interface daunting." To say the least.

    10. Re:You underestimate Bill Gates by noamsml · · Score: 1

      Sorry if I misread you, but I get annoyed when someone claims that 'nerds', 'socially implaired' or 'isolated people' are objectively inferior to them, and/or that sexual activity is the only way to measure one's achivement. That's complete an utter nonsense. Fuck, I am the only one who can give a meaning and a goal to my life, and I don't care what the rest of society thinks I should achive, it's my life.

    11. Re:You underestimate Bill Gates by vsprintf · · Score: 1

      Sorry if I misread you, but I get annoyed when someone claims that 'nerds', 'socially implaired' or 'isolated people' are objectively inferior to them, and/or that sexual activity is the only way to measure one's achivement. That's complete an utter nonsense.

      I said nothing about sexual activity or any related inferiority. A sense of humor is vital in social interaction, and as distasteful as it may be to some (I lean that way myself), that interaction is necessary to get most things done.

      Fuck, I am the only one who can give a meaning and a goal to my life . . .

      I hear you, but you're wrong. I once felt the same, but there is someone else out there in the world who can give a new meaning and a new goal to your life. It will happen. It doesn't change your direction - it only adds new dimensions. A ready sense of humor is your best ally in finding that person.

    12. Re:You underestimate Bill Gates by vsprintf · · Score: 1

      Now you're all hot and bothered... sweaty, shining a little in some places... fogging up windows, you get my drift.

      Hmm. . . That sounds like a quote that I should be able to remember. You taunt me.

    13. Re:You underestimate Bill Gates by Anonymous Coward · · Score: 0
      Learn to spell, loser.

      (I feel better now!)

  92. Comments on others' comments by vectorian798 · · Score: 3, Insightful

    Security issues are irrelevant in a lot of cases. Scientific computing isn't done on computers attached to the internet...it is done on intranets consisting of specialized hardware streamlined for the needs of HPC. Most HPC programs don't even attach themselves to ethernet networks, but rather to things like Myrinet (bypass OS calls to reduce overhead GREATLY) that are intended for HPC. Being DDoS'd, or having a 'zombie cluster' etc are not really issues here.

    I think the advantage of a MS solution might be ease-of-use, especially in server clusters that are up for hire (that is, up for timesharing). If you are some group performing research that requries lots of power but aren't focused in a CS-related field, you may not have the resources to go use the (often arcane) parallel (MPI) debuggers etc. and churn out a top-grade program for a supercomputer. An MS solution might indeed be cheaper OVERALL because of time-to-solution (time = money). Let's face it, VS.NET is a dream to code in - compared to other well-featured IDE's like Eclipse, it is light-weight, easier to use (Eclipse has major bloat issues), etc. So who knows - as the article mentions, it might indeed become part of an end-to-end scientific process, where the computational parts seamlessly fit in.

    Furthermore, everyone who is talking about licenses per processor are not thinking properly...do you really think they would achieve penetration with the barrier to using the software so high? Of course not! Instead of speculating negatively, let's just wait and see what the licensing programs are when the product is released.

    My 2 cents

    1. Re:Comments on others' comments by Anonymous Coward · · Score: 1, Informative

      Tell that security part to the NSF, Stanford or a bunch of other
      HPC cluster sites that were "owned" last year thru a security issue.

      Scientific computing isn't hooked into the Internet ?? get real, it
      most certainly is. Unclassified sites often support users across
      the globe. For those HPC sites hooked into the Internet that want
      to play fast and loose with security, be my guest. Just remember
      when your funding dries up because you were "hacked" or lost
      some researchers data that "security isn't an issue".

    2. Re:Comments on others' comments by MaskedSlacker · · Score: 1

      They acheived domination with the barriers that high so far. Comparable software to what I use every single day in Linux would cost me $10,000 in windows. And I am one person, these are my personal systems.

    3. Re:Comments on others' comments by lheal · · Score: 1
      If you are some group performing research that requries lots of power but aren't focused in a CS-related field, you may not have the resources to go use the (often arcane) parallel (MPI) debuggers etc. and churn out a top-grade program for a supercomputer. An MS solution might indeed be cheaper ...

      If you lack in-house expertise at implementing parallel algorithms (or parallelizing yours), then you have to shop around for someone who can help you. As long as you need that kind of help, you probably won't be buying something, you'll be using someone else's cluster.

      Looking at it another way, if you can't learn MPI you have no business running a cluster. It won't do you any good.

      --
      Raise your children as if you were teaching them to raise your grandchildren, because you are.
    4. Re:Comments on others' comments by NatteringNabob · · Score: 1

      You must have gotten a different version of VS.NET then I got through the student discount program at my Univ. I brirfly had it installed on my laptop (dual boot XP and FC3). I paid $15 and it would have been a bargain at 1/2 the price. It sucks up an ungodly amount of disk space, wastes tons of screen real estate on eye candy (just what I need in an IDE!) and is not in any sense 'light weight'. I uninstalled it as soon as I needed the space for something useful and went back to booting Linux, coding in emacs, and debugging in Netbeans 4.1. There is a huge need for a light weight IDE for Java and .NET programming. VS.NET isn't it.

    5. Re:Comments on others' comments by Anonymous Coward · · Score: 0
      Instead of speculating negatively, let's just wait and see what the licensing programs are when the product is released.

      No matter what the licensing terms, it is hard for me to face the prospect of HPC Windows positively when I've just successfully purged Microsoft software from my life. Microsoft are doing this for bragging rights, if a cluster helps finds a cure for cancer then Gates wants to make damn sure it is Microsofts name in the press release. [OT] Interestingly, all the classic narcissistic traits apply equally to both Gates the individual and Microsoft the corporation.

      --
      Never attribute to malice that which you can attribute to a mental illness.

    6. Re:Comments on others' comments by Cyno · · Score: 1

      let's just wait and see what the licensing programs are when the product is released.

      Yeah, you go do that.

      I couldn't care less about this product.

      I've already got one, you see..

      Why should I support Microsoft or even take the time to look at their products and read their licenses? They are irrelevant today. They chose to be this way.

    7. Re:Comments on others' comments by Salis · · Score: 1

      Security issues are irrelevant in a lot of cases. Scientific computing isn't done on computers attached to the internet...it is done on intranets consisting of specialized hardware streamlined for the needs of HPC. Most HPC programs don't even attach themselves to ethernet networks, but rather to things like Myrinet (bypass OS calls to reduce overhead GREATLY) that are intended for HPC. Being DDoS'd, or having a 'zombie cluster' etc are not really issues here.

      Scientific computers are _often_ connected to the internet. How else do scientists/engineers connect to them? And hacking does happen. (It happened at my institution.)

      The connection is usually over ssh. It's secure, but not perfect.

      --
      Favorite /. tagline: "On the eighth day, God created FORTRAN." And it was good.
    8. Re:Comments on others' comments by Dan_Bercell · · Score: 1

      Then why even post under MS threads? You obviously are interested in what MS is doing otherwise you wouldnt be reading about it

    9. Re:Comments on others' comments by Cyno · · Score: 1

      I do care about Microsoft. I worked at Netscape. Microsoft is a convicted monopoly and their punishment did not fit the crime.

      Why do you care so much for Microsoft that you want to try out their unreleased products, in a field in which they have no historical expertise, and free alternatives (that enforce liberty and fairness) already exist? Do you care more about a product's ease of use than about fair business practices? And what about the liberty of its end users?

      So I take it you're for free market economics and survival of the fittest? Me too.

  93. Typo alert! by bombshelter13 · · Score: 0, Flamebait

    >Microsoft Competes In Supercomputer Market

    I think you mispelled 'Fails To Compete'.

  94. you got it by zogger · · Score: 1

    Bragging rights, mindshare, the integrated environment and all sorts of various marketing reasons. MS wants to be the computer system for all uses and users from gaming consoles and cellphones to super computers. And they certainly have enough money to try. I haven't looked, but bet their advertising budget is higher than the gross income for all the linux distros and apps. And if it isn't they could make it so tomorrow and not break sweat. Not saying it is wise, or warranted, but they have the proven ability to extract cash from society like no other single company ever has before.

  95. you misspelled poses by wardk · · Score: 0, Flamebait

    "poses" is spelled

    P-O-S-E-S

    not C-O-M-P-E-T-E- S

    that's an entirely different word, and I think the meanings are still different too.

  96. Re:Claiming? by st1d · · Score: 2, Insightful

    >>When I bring hardware home, it works! Try getting those wireless usb devices working in Linux. Possible, sure! Do you just plug it in the slot and put in the CD and it works, not *even* close. Audio on Linux? You've got to be fucking kidding me. What fantasy world does the poster live in where Linux is easier to use than Windows?

    What fantasy world do you live in where installing Windows and several additional third-party drivers is "easy"? At least with linux, you can generally avoid most of the driver installs -- they are included in the kernel, and are found automatically, but you might need to tweak a couple things before they work (or work perfectly). I'm amazed when I watch friends talk about Linux being difficult to use, meanwhile they'll spend hours reloading windows (and drivers, etc.) every couple months, if not more often.

    More frustrating, if that's the right word, is that people complain about how "difficult" it is to research/make fixes for Linux, but don't even try to fix problems in Windows, opting for a reinstall instead. Kind of an apples and oranges thing, when you think about it.

    --
    Microsoft has just released their much anticipated hands-free cordless mouse. Warning, it may hurt a little at first.
  97. Serious computing with Windows? by Anonymous Coward · · Score: 0

    This Windows cluster reminds me of the TK-2000, a Brazilian Apple IIc clone that wasn't even compatible with the IIc. It was marketed to professionals and small businesses as "a business computer with a professional keyboard". On the other hand, the keyboard had two big and prominent keys labeled FIRE. So much for professionalism.

  98. oh wow! by Eric+Damron · · Score: 1

    A cluster of Windows machines! That crash should be spectacular!

    --
    The race isn't always to the swift... but that's the way to bet!
  99. No troll jokes here by genner · · Score: 2, Insightful

    Finally a slashdot story where no one can possible imagine a beowolf cluster of these.

  100. Microsoft Lawyers ... by Anonymous Coward · · Score: 0

    Microsoft's NEW Cluster EULA:

    Cluster Home Edition: 5 node max
    Cluster Professional Edition: 10 node max
    Cluster Server Edition: 10 nodes starting at $10,000. $1000 for each additional node.

    A bargin at any price...

  101. Purse Strings? by seven+of+five · · Score: 3, Insightful

    Regardless of what the folks who actually use the supercomputer want, there's always this administrator who signs off on the purchase who will say 'Windows, huh? Great. Now we can have one support contract that covers everything!'. The M$ Sales Rep takes him/her out to a couple fancy lunches and comes back with a signed contract.

  102. It has always been thus by A+nonymous+Coward · · Score: 1

    FORTRAN was invented because people spent too much time debugging programs on those expensive computers. If a FORTRAN program was half as efficient as machine code, but the machine code program spent 90% of its time being developed and debugged, the FORTRAN program did roughly five times the work (50% vs 10%).

    Now whether Bill's wet dreams work out is another question altogether. He may get a few piddling sales because of heavy discounts for bragging rights, but centers with gazillion node supercomputers can easily devote a few of those nodes for development and leave 99% of them for the finished programs. Visual Basic won't have much to offer in that case.

    1. Re:It has always been thus by Anonymous Coward · · Score: 0
      centers with gazillion node supercomputers can easily devote a few of those nodes for development and leave 99% of them for the finished programs.

      It's not about the computing resources needed for development - that's pretty trivial. It's the human resources. If it takes a week to run on the supercomputer and 6 months to write the code, you may be better off spending you resources on the development end rather then the runtime end.

  103. Windows Compute Cluster Server 2003 ... by tonymus · · Score: 3, Funny

    ...when you need to run MS Word really, really fast.

  104. Wow... by Anonymous Coward · · Score: 0

    Does this mean that he is admitting that Linux is more powerful than Windows?

  105. The real point of this move by NewIntellectual · · Score: 0
    I haven't seen any other comments address the actual environment Microsoft is addressing. This is from a New York Times article on the move:

    The company's Windows Computer Cluster Server 2003 software is scheduled to become available in the first half of next year and is intended to give scientists and engineers a simple way to gain high-performance computing from their existing Microsoft desktop computers.


    So, this is the same sort of distributed computing used for years for SETI@Home, etc., presumably with a nicer interface and software to facilitate scheduling remote processes with pooled results.
  106. Bleary eyed by minion · · Score: 1

    ... I just woke up and checked Slashdot... Have I been sleeping so long its now April Fool's Day?

    --

    -- If we don't stand up for our rights, now, there will be no right to stand up for them later.
  107. Day Late by Anonymous Coward · · Score: 0

    M$ Always a day late dollar sho... wait a min..

  108. Re:Metrics: BSPS by vettemph · · Score: 1

    BS Per Second?

    --
    The government which is strong enough to protect you from everything is strong enough to take everything from you.
  109. Structure of the stack by BigFootApe · · Score: 2, Interesting

    There are a few features which Microsoft would have to implement on any OS targetted at cluster computing. Many are not directly within their control.

    First, and most important for users, what would be the APIs provided. Would Microsoft package MPI? PVM? Would they use a proprietary technology? XML based technologies are way too heavy for this application.

    Second, what interconnect transports would be provided? VIA, Globus, IB, good old stinky rsh encapsulation? What about independent vendors like Myricom and Dolphin? Would these companies be willing to support a substantially different architecture? Would there be enough customer demand for them to support Compute Cluster Server at the outset (MPICH-GM is old old old for Windows, Dolphonics and Scali are pretty well exclusively LINUX)?

    Third, what software will Microsoft be providing for remote batch management? You'd need a secure remote shell, good scripting functionality, non-GUI device management, etc.

    Lastly, how suitable is the NT kernel to doing this sort of work? VMS was ahead of basically everyone when it came to clustering technology, yet _nobody_ uses or used it for parallel processing. What are the lessons that can be applied to NT?

    There are a few clusters built on NT, but most of the ones mentioned on the Beowulf mailing list (and they are few) are networks of workstations with CONDOR installed which do double duty as computer clusters at night.

  110. "Easy to Use? Who cares?" by Mente · · Score: 2, Insightful

    Who cares if a supercomputer OS is easy to use? The only thing people that use supercomputers care about is speed. Other things supercomputer users care about is compiler efficiency (again for speed.) I don't see Visual Studio cranking out the efficient executables. I'm also quite sure the Joe Schmoe Ph.D in CS, CEE and Physics does not care if an eight year old can navagate the user interface. But he does care whether or not it takes the computer 2 weeks, or 3 months to model galaxy formation. Especially when he knows that there are 7 other teams that just got the same new information from the same conference and are all racing to publish their newest findings.

  111. Should that headline say... by Liam+Slider · · Score: 1

    Microsoft Doesn't Compete in Supercomputer Market?

  112. this is it... by nappingcracker · · Score: 1
    • xbox 360 built with multiple big-bad-multi-cell(tm) processors with lots of memory bus
    • Microsoft launches xbox 360 with xbox live for free
    • Microsoft announces that it is getting into supercomputing

    [lots of xboxes] + [freenetwork that everyone with broadband is likely to use] + [redmond supercomputing r&d] = skynet

    Sure, there would be lots of latency, but with the projected install base...makes for one bbmf(tm) of a distributed net. Each of the boxes will be nearly identical and will includeh an OS that could be tweaked remotely and across all nodes without much hassle, sounds good for engineering a distributed net, yea?

    How will one know that when updateing a live account, or downloading new content you are not sending the next piece of the puzzle? What if the games include pieces of the program distributed, compiled and redistributed by the live master control like the Joker's happy fun chemicals in the first Burton/Batman movie? They control the sdk, the compiler, the box, the final review of the code, the networking, why not use all the gear to compute big stuff(tm)? Let consumers buy and build your Xmillion node grid? Sic!

    Yes, that last bit (or the whole bit) makes some serious and likely fubar assumptions about how/if/what the whole idea works. Yes, this is another cracked out late night post of consipiracys and ill-informed crap. However, I am sure that more level headed chaps could trip out on the idea and see if it is technically possible. And if something is technically possible...

    [mumble]...ill trip on that trying to sleep tonight...mislead about foil hats...skynet in my living room...robot reservation...[yawn]
    --
    |plastic....or gasoline?|
  113. ease of programming for, ease of using by mangu · · Score: 1
    So there is growing interest in the notion of "time to solution" as a combination of ease of programming for, ease of using, and of course running a data set on the machine.


    Sure, you got it quite right. A computer that's useful for a scientist has tools that can find the eigenvectors of a matrix, calculate the positions of planets in the solar system, solve the Navier-Stokes equation to plot the shock waves around a supersonic wing. Exactly the kind of problems which are so easy to solve using Microsoft Windows...

  114. Better than Moore's law? by DonGar · · Score: 2, Interesting

    1991 to today is 14 years, or 9.33 cycles or Moore's law. 40 Mill, halved 10 times is $39,062.50. Since the article is talking about hardware for under $4k, the price is about a tenth of what was predicted.

    I'm not drawing any conclusions, just pointing this out.

    --
    plus-good, double-plus-good
    1. Re:Better than Moore's law? by Anonymous Coward · · Score: 0

      It's actually better than that, because you haven't considered inflation. $40 million of 1991 dollars is worth much more than $40 million of 2006 dollars.

    2. Re:Better than Moore's law? by photon317 · · Score: 1


      Moore's Law is not neccesarily about what the maximum gflops/dollar is over time (actually, who really knows what it's really about for sure... see: http://en.wikipedia.org/wiki/Moore's_Law), and on top of that, a Cray Y-MP was probably never the best gflops/dollar package, even in its heyday. The old Crays were more about maximum gflops, period. There was probably a low-end processor at the time that cranked out more gflops/dollar, but commodity clusters of cheap processors were virtually unheard of at the time.

      --
      11*43+456^2
    3. Re:Better than Moore's law? by Anonymous Coward · · Score: 0

      Two comments:

      1. Cray was always in the "top of the line, non-commodity" market. You pay a premium for that. Once a technology goes from the non-commodity market into the mainstream you'll get a big blip in the bang/buck ratio

      2. Mainstream FLOPS performance has outpaced Moore's law over that period simply because it had been *so* bad. Hell, in 1991 the first x86 processor with an on-chip FPU (486DX) was only a couple years old. In those days even low-end RISC workstations just blew the doors off any PC when it came to FP.

      Now x86 may not be at the top of the FLOPS game but it's in the same ballpark as the leaders. Therefore you can get great FLOPS/$ out of normal chips.

      So in 1991 in order to build a FLOPS-monster you had no choice but to use non-commodity parts. Now you can do it by using the same CPU/motherboards/RAM/etc that teenagers use to play video games.

    4. Re:Better than Moore's law? by Anonymous Coward · · Score: 0

      Although the processing power maybe equal to a 1991 Cray, the memory bandwidth (which has a major impact on throughput) is nowhere near that of a 1991 Cray. You also need to remember that Cray's are vector processors and very well suited to certain operations, but may not appear so high on the benchmarks which are more generic programs. If you wanted to do a physics simulation (pretty common use of a supercomputer) you may find that they processing speed is nowhere near equivalent. Think of it like comparing a graphics card processor to a general purpose CPU and asking both of them to perform a non-graphics related benchmark.

  115. You're all missing the point by Anonymous Coward · · Score: 1, Insightful

    I think everyone here is missing the point.

    This software is for biologists, chemists, and other scientists who don't have a damned clue what a shell script is, much less NFS, etc.

    I work to support local users on a school supercomputer, and they don't have a clue of what's going on 90% of the time.

    All they know is that they want software X run with parameter Y1-Y100 done Z number of times. They don't want to spend precious research time figuring out the computer systems.

    This software is for a biologist who wants to buy a small cluster, plug it in, install his software, and run. No configuration, zero administration. Eliminating the need for the computer illiterate to have a sysadmin to run a cluster; THAT is the power of Microsoft's software.

    That said, as a computer literate person, I wouldn't dream of installing Windows on a cluster; Linux and the OSG is the only way to go.

  116. sorry gates by cood · · Score: 1

    but you're in for a rude awakening.

    --
    Average is dumb :)
  117. Windows Compute Cluster Server? by Anonymous Coward · · Score: 0

    does not compute. does not compute. does not compute.

  118. M$ should not venture from the desktop world by elstonj · · Score: 1

    How do you get supercomputer performance from something that can't run without running a GUI? 50% of your available processor time is spent running processes that are NEVER used. Oh yeah - I forgot - everyone administers their clusters by hooking a keyboard, mouse, and monitor to each node and clicking start->settings->control panel

  119. Imagine... by formant · · Score: 1

    ...a beowolf cluster of shit !!

  120. I'm sure olde Billy G has alot to offer attendees by Locutus · · Score: 1

    NOT!

    How these guys buy their way into conferences they have no right being at is beyond me. Let alone marketing...I mean speaking/presenting at them. I pity the fools who have to attend Gates speach because their bosses told them to.

    LoB

    --
    "Anyone who stands out in the middle of a road looks like roadkill to me." --Linus
  121. I humbly apologize by wardk · · Score: 1

    oh, I am sorry.

    I meant, I am in awe regarding this pivital moment in computing history

    we will all want to remember where we were, when we heard about this momentus occasion.

  122. Point-and-click supercomputing by Anonymous Coward · · Score: 0

    I can see it now...

    Bill Gates gets up on stage, pulls up the latest edition of Visual Basic .SUPERCOMPUTE, and proceeds to throw together a quick application, just to show how easy it is: drop in an atmospheric grid-lattice form, stick in a time-series forward-differencing control here, quick call to a 3D visualization class there, top it off with a seasoning of atomic isotope models, and he's got a complete working app that simulates a nuclear bomb explosion, while disproving global warming at the same time, all in the space of 10 minutes.

  123. Blue Gene and its progeny will eat this alive. by Tetrachromatic · · Score: 1

    IBM has proven with their custom Blue Gene OS that a minimal, focused and efficient operating system trumps the notorious bloat and pack-in that every modern Microsoft OS has been known for. Try and think back to the last time you've described any product in the Windows Family as "lean". IBM has also been very keen on Linux, tailoring it to Blue Gene's I/O nodes. Therein lies the beauty of custom operating systems and highly customizable operating systems, two arenas where Microsoft just can't compete. You can get what you want with as little overhead as possible.

  124. I wonder what a super-corrupted registry is like? by Anonymous Coward · · Score: 0

    I wonder what a super-crash is like? Or a super-blue screen? Or a super-corrupted registry? Or super-bloat?

  125. Anyone paying attention to huge databases? by Anonymous Coward · · Score: 0

    HPC can perform incredible analytics on databases... Think more along the lines of Clippy saying, "Would you like to analyze performance of all the companies in the Pacific Rim only or the entire world?"

    Or perhaps distributed, server side code for web services??

  126. Microsoft is looking to the future by Anonymous Coward · · Score: 1, Insightful

    Microsoft aren't stupid or paranoid, or at least they can see what's bleedingly obvious, which is that Windows can't compete in the supercomputer space. Not now, but what of the future? CPU designers, with shrinking geometries, are running out of room to manoeuvre, so are looking to parallelism for future growth. Windows needs to get into the cluster space now, otherwise in 5 years time, they'll find themselves out of the market.

  127. "Windows" is silly name for HPC OS by Bandraginus · · Score: 1
    How could anyone take "Windows" seriously in the HPC environment when having "Windows" is at odds with the very paradigm of clusters. I don't know of any supercomputer that has a head per node on which to display all these Windows.

    Having an OS named after a GUI paradigm is just a little silly in a HPC environment.

    Maybe it's time for Microsoft to come up with a new name for its OS in this space?

  128. Lil Blue by Anonymous Coward · · Score: 0

    I suppose they couldn't buy up IBM to get hold of the title could they and find out how to design serverware by finding loads of machines running Unix and Linux and st...

    Nahhh to fanciful.

    What I can't understand is why they bother. The bloody servers they have now can't even handle their own groups. After all these years you still can't post from Word to their own communities without looking a right NOBR

  129. Opportunity window : OS Paradigm shift ! by Anonymous Coward · · Score: 1, Interesting

    Okay, ill admit to a round of buzzword bingo :

    There is an OS paradigm shift looming.

    With the incorporation of "virtualization features" in the next round of x86 cores from both intel and AMD, the two-layer software world (OS / userland) of the last 40 years is complemented by a third "virtualization layer" at the SW/HW boundary.

    This means that with the next OS-Upgrade that supports this virtualization, every corporate desktop is able to run several instances of one or several OSs with a high degree of isolation.

    This mean Joe Beancounter at GM Headquarters can pop a Britney Spears CD from sony with a rootkit into a CD-Drive, catch a virus and begin to spread SPAM through GMs fat pipe to the world, while a tiny slice of the crash test simulation for the new Cadillac is running on the spare cycles on his desktop completely undangered from virtual bluescreens and reboots.

    Of course, Linux could do this just as easily as Windows, but how many identically configured linux desktops does GM run ?

    When GM rolls out Windows Vista, in, say, three to five years time to 100.000 Desktops, they get a 100.000 node cluster thrown in for free.

    It is available with 80% capacity 8 hours a day and 100% capacity for 16 hours a day plus weekends.

    Okay, it is very loosely coupled, and thus much less effective than a dedicated cluster, but I guess that for a lot of applications it would be more powerful than a 20.000 node dedicated cluster.

    So this would be a $40 million machine for basically free.

    All GM would have to do is start porting one of its apps to the new architecture now.

    And of course the savings would look even better for underfunded universities.

    I think that if "Windows HPC OS" is included in the "virtualization support" in each Windows Vista OS, and they do a decent job of implementing the application deployment and data distribution, Microsoft stands a good chance of getting a hold in the market.

  130. Hmmmh..... by KwKSilver · · Score: 1

    Win-don'ts... Doze? ... I give up.

    --
    If you want your life to be different, live it differently.
  131. you rock! by ine8181 · · Score: 1

    you rock!

  132. No! You have underestimated Bill Gates by Anonymous Coward · · Score: 0

    They want to create a new market targeted toward a different kind of people

    Windows has, believe it or not, already got a large customer that would be able to use his malware. The US Navy. I remember reading about it ages ago. (No links though.)

    I think their OS is going to be called "Portholes". Apparently they got the contract when George Bush found out about Windows and Open Source.

    He thought the ships would be safer with them closed.

  133. HAHA HAHA HAHA!!!! by Anonymous Coward · · Score: 0

    "Microsoft Competes In Supercomputer Market"

    Didn't know it was april yet...

  134. Anyone Remember Wolfpack? by asink · · Score: 2, Insightful

    I do. This isn't microsoft's first try at this. I expect similar results

    --
    "Hex, Bugs, and Rockn'Roll"
  135. Va Tech by 1336.5 · · Score: 0

    Supercomputing Cluster....all Apple G5's

    get.a.clue

  136. First? by Impy+the+Impiuos+Imp · · Score: 1

    > in their first attempt to compete in the supercomputer OS market.

    David Spade: I liked this OS better the first time...when it was called Windows NT.

    --
    (-1: Post disagrees with my already-settled worldview) is not a valid mod option.
  137. Or is that... by Anonymous Coward · · Score: 0

    Microsoft Competes in Supermarket Computer...

  138. I just can see it now.... by Fengpost · · Score: 1

    Ballmer throws a chair across the room and says

    "I am going to bury Blue Gene, I have done it before......"

    MS needs to get a clue.

    --
    The purpose of writing is to inflate weak ideas, obscure poor reasoning, and inhibit clarity....Calvin
  139. Gates says "easier to use than Linux" by 1336.5 · · Score: 0

    I know people think Gates is a genius in some circles but does it take a rocket scientist to throw a brick through a glass window in the middle of a riot?

    ANYTHING is easier to use than linux. Good job Gates, way to make a revolutionizing prediction, or as I would rather think about it - way to state the obvious there jackass.

    mod points pls?

    "Some people are like slinkies. They are really good for nothing, but still bring a smile to your face when pushed down a flight of stairs.

    -1336.5

  140. from the Reg article: by cosminn · · Score: 1

    "What we see as a key trend here is that we will have supercomputers of all sizes, including ones that will cost less than $10,000 and be able to sit at your desk or in a department," Gates said.

    that's so much crap, the trend is to have thin clients, to have WEB apps instead of bulky software Let's keep the supercomputers for the scientists and the desktops for the average consumer...

    I mean, if the trend is for supercomputers under my desk, why the fuck is Microsoft releasing Windows and Office LIVE ? ;)

  141. Re:How much would it cost to be wrong by Anonymous Coward · · Score: 0

    This: And that means that those universities are going to have rewrite their custom apps at great expense. I can't see it. I simply can't see how MS can reasonably strong arm anyone, corporate, governmental or university, into taking on a supercomputer variant of Windows.

    And even if MS were so compelled, what really is in it for Redmond? It's probably the smallest market in the world, with a customer base measured in the thousands,
    is the kind of thinking that put IBM where it is today.

    One day everyone will want a deskbesides PCC. They were about one and a half times the height and twice the base area of a standard ATX case ICIRC the picture of the one I was last drooling over. But that was many moons ago.

    They'll be a lot smaller by now. A lot. And probably even more powerful and cheaper too. They may even have cured the bloody awful noise on it.

    OOH! And they weren't utilising physics and graphics card potentials then, too neither...

    "Ooo..mmmm clusters..."

  142. You might want to read the article. by porkchop_d_clown · · Score: 1

    Yes, this product will include MPI and a suite of cluster management tools.

    As someone who actually works in this area, I don't see this as much weirder than Apple's offering. In particular since small scientific computing clusters tend to run one job at a time, Windows' relatively poor multiprocessing capabilities won't be such a loss.

    I'm not a big Windows fan (I run Linux and OS X at home and I write Linux drivers for a living) but I can see where a Windows oriented IT department would see this sort of thing as a win - they can set up the kind of cluster that their internal customers want without having to learn one of them there communist operating systems.

    Actually the more I think about this the more I see it as a response to Apple's Xgrid and Workgroup clustering solutions than an attempt to break into the Top 500 list.

    1. Re:You might want to read the article. by BigFootApe · · Score: 1

      I did read the article. It mentions that Microsoft is building a batch dispatching program and "other utilities". Reading another article (linked to by the one posted), Microsoft will also have a profiler and an MPI library. Goody. What about my other questions?

      I actually don't understand how Apple's offerings make any sense. On a desktop system, paying some extra bucks for ease of use makes sense. All the polish will tend to save you time over the long run, as well as lowering the proficiency bar for your employees. Same thing with a server sitting in a DMZ, where updates are rather more frequent. A cluster is more like an embedded system. You set it up once and forget it for a few years (I should know, I've done it a time or two).

  143. Re:Metrics: BSPS by darkmeridian · · Score: 1

    I thought BSPS was a measure of Microsoft's public relations staff. Like the BSPS goes up whenever there's an astroturf TCO article.

    --
    A NYC lawyer blogs. http://www.chuangblog.com/
  144. As Snagglepuss would say... by RoadWarriorX · · Score: 1

    With Linux, you don't even have to install the OS on the hard drive. It can run off of a CD.

    As Snagglepuss would say:

    It can run off a network, even
    Time for Windows to exit, stage left!

  145. Re:You misunderestimate Bill Gates by Foofoobar · · Score: 2, Insightful

    Yes, and my mom has been saying for the longest time how she wants to operate a supercomputer. Several mom and pops can now own and operate super computers too.

    Again, i think the original writer was correct. Supercomputers are for enterprises and government agencies. Even some medium size businesses would have a cluster. But they all have a fulltime sys admin. And if the extent of that sys admins abilities is just a MCSE cert, then god help your company when the latest exploit hits.

    Also, keep in mind that if this is using a Windows interface (rather than a shell prompt), it is using easily 3X the resources that a Linux machine without X-windows installed would.

    Finally, since when has a first version of of any Microsoft product worked great?? There is a reason why everyone waits til the third edition to really adopt a Microsoft product. But in this case, they are fighting an uphill battle against something that is more scalable, faster, more stable and cheaper by far. Plus it's been tested time and time again.

    If Windows could handle the load (and if Microsoft was worth it's salt as an server OS), they would stop using Linux to do all their load bearing. The day Windows cluster appears in the top 10 supercomputers is the day George Bush comes out of the closet, marries Dick Cheney and decides to become an atheist.

    --
    This is my sig. There are many like it but this one is mine.
  146. Customizeability! by Anonymous Coward · · Score: 0

    K Menu -> Control Center -> Peripherals -> Mouse -> Single Click to open Files and folders.

    I've always liked the fact that KDE is so customizable. Right now, I have 4 panels, and Screen Fliping enabled (for virtual desktops), and both of those features are unique to linux, so I dont know why people accuse KDE of having no original ideas, and for cloning windows.

    1. Re:Customizeability! by vsprintf · · Score: 1

      K Menu -> Control Center -> Peripherals -> Mouse -> Single Click to open Files and folders.

      Thanks, following that lead, on my distro, it's more like Start Menu -> System -> Configuration -> KDE -> Peripherals -> Mouse -> Single Click to open Files and folders. That doesn't seem intuitive to me, since I expected that control to be under window (not mouse) behavior, since you can select icons using a mouse or keyboard. In any case, I still think KDE should use default controls that make the most sense, not the ones that mimic Windows.

      I dont know why people accuse KDE of having no original ideas, and for cloning windows.

      It's not that they don't have original and good ideas, it's that they hide those ideas under layers of menus and accept the Windows way by default. I can understand they are trying to be familiar for Windows users that are making the transition, but that is just hiding the innovative parts from most of us regular users. I think it's time for KDE to concentrate on KDE and do it right by default. Provide options for people who like the MS interface if they want to select it (or give Linux users an easy one- or two-click way to a cleaner interface). Let Microsoft play catch-up.

    2. Re:Customizeability! by Nasarius · · Score: 1
      It's not that they don't have original and good ideas, it's that they hide those ideas under layers of menus and accept the Windows way by default.

      Wrong. Your distro is obviously screwing with the defaults. Install KDE from scratch, and you get the single-click to run by default. You also get some strange stuff, like double-clicking the title bar will "fade" the window.

      --
      LOAD "SIG",8,1
    3. Re:Customizeability! by vsprintf · · Score: 1

      Wrong. Your distro is obviously screwing with the defaults. Install KDE from scratch, and you get the single-click to run by default.

      I will take your word for it. I have tried to update KDE in the past and completely hosed everything. There tends to be Qt and/or gcc/glibc dependencies that affect everything.

    4. Re:Customizeability! by Anonymous Coward · · Score: 0

      There tends to be Qt and/or gcc/glibc dependencies that affect everything. Why not let your package manager resolve them for you automatically?

    5. Re:Customizeability! by vsprintf · · Score: 1

      Why not let your package manager resolve them for you automatically?

      When there are no blessed packages that cover all the dependencies for the latest KDE release? What are you smoking?

  147. Oh my... by Anonymous Coward · · Score: 0

    /cue music

    Everyone sing along
    It's the end of the world as we know it...

    I'd prefer a cluster of ColecoVisions!

  148. ctrl-alt-del by Golden_Eternity · · Score: 1

    One would think that needing to reboot your supercomputer every couple days might not be the best selling point.

    1. Re:ctrl-alt-del by Compumyst · · Score: 1

      You might want to check your math. Take, for example, 1 week:
      7 days/week X 24 hours/day X 60 minutes/hour = 10080 minutes.
      So if this cluster runs at about 1000x as fast as a standard Windoze computer, then you'll need to reboot about every 10 minutes.

      On the other hand, it would probably only take about 1-2 second to boot up (excluding the post'ing)

      --
      What's done's in the past, forever shall last.
      Work is work; life is life; fair is not!
  149. This is about mindshare. by BluSteel · · Score: 1

    To maintain the monopoly, Microsoft must maintain their mindshare. "Most of the world's supercomputers run Linux" raises questions in John Q's mind, and might sew the seed of future rebellion. Linux must have no stronghold ANYWHERE or you can't keep the masses ignorant.

  150. The best rendered 3-D Pipes around by Anonymous Coward · · Score: 0

    What kind of performance is your Microsoft Cluster really going to get when all your nodes are running a 3D Pipes screen saver...

    The problem is, most of the cluster nodes are likely absent graphics cards, keyboards, and mice. So how do you even boot/configure your "Microsoft Cluster"?

  151. The people that Really Benefit by megarich · · Score: 1

    A windows supercluster? You do know who this is going to make very happy? Yup the porn,online gambling and the russian mob spyware/malware industry.

  152. They might want to write a Fortran compiler by andy314159pi · · Score: 2, Insightful

    If Windows wants to get in the HPC market, they should start by offering a fortran compiler. Lahey has one for Windows, but for a native UNIX user, the instructions on how to use it on Windows were just about incomprehensible. Strangely, for HPC, Linux, AIX and other UNIX systems are probably much easier to use. And I doubt windows binaries would be as fast anyhow.

  153. Re:what really happened by floating+eye · · Score: 4, Funny

    They spent billions to make a new Windows which turned out to need a supercomputer to run, and now they are desperately trying to find customers.

  154. Paperclip!! by twoblink · · Score: 0

    Great! Now you can be annoyed by the paperclip 1000000x faster!!

  155. Re:Metrics: BSPS by Johnny+Mozzarella · · Score: 1

    Windows Compute Cluster Server 2003 will also give "GigaFLOP" a new meaning.

  156. Self interest by jdunck · · Score: 1

    Why would MS want to market a supercomputer OS?

    Because they had to build one anyway; they may as well see if they can sell it.

    Google is kicking their asses in the cloud-of-services arena, and its doing that on the basis of a 200,000 server farm.

    MS is looking to defray their catch-up costs.

  157. MS already tried this with Unisys ... by Salis · · Score: 1

    MS had a joint venture with Unisys a while back. They loaded a 32 proc cluster of Itanium2's with the Windoze server OS and _donated_ it to my institution for Bioinformatics research. Well, the machine sat for the longest time unused because _none_ of the HPC people in my _entire_ University wanted to use it. Eventually, they partitioned the system and loaded Linux using 16 of the procs. And...people started using it.

    I think it's prediction of what's to come.

    --
    Favorite /. tagline: "On the eighth day, God created FORTRAN." And it was good.
  158. Re:The difference is bullshite by jerkyjunkmail · · Score: 1

    I call bullshite. Napster relies heavily on DRM'ed WMA files. Their website doesn't have much on it regarding the real restrictions. Maybe some small indie labels allow them to sell un-DRM'ed MP3's but all the big pop songs like gwen stefani or coldplay are DRM'ed WMA files and NOT unencumbered MP3's which explains why iPod isn't on the list of supported players Apple didn't license WMA support for the iPod in spite of it being capable to support the DRM'ed WMAs. you might workaround the WMA DRM by burning them to a CD and then re-ripping the CD to standard MP3's but they are not selling or renting you unprotected files. Re-encoding/ripping is possibly against the EULA.

    jerky

    --

    --
    What is pirate software? Software for inventory of stolen treasure?
  159. Re:The difference is bullshite by NitsujTPU · · Score: 1

    You're uninformed. Click the "purchase" button on Napster. Then, click the "purchase" button on iTunes, and tell me what kinds of files you have.

    I play my napster files on a device that was not designed to support napster. The DRM is only for the services where you do not purchase individual songs.

  160. FLOPS by kai.chan · · Score: 1

    And running at a speed of Always-FLOPS.

  161. The target audience by Anonymous Coward · · Score: 2, Insightful

    This stuff is marketed to the people who say:

    Our windows-based programs are too slow. Don't tell me we need to go to programs that are written more efficiently, or that there would be an overall cheaper and superior non-windows-based alternative: I don't want to hear it. Change frightens me. Just give me more power.

    I see this mentality all the time. Here's an example from school I remember. This one student would sit and take up multiple windows computers in the computer room to run Matlab. He did this for months.

    Did he want to hear that he could invest some time and recode in a fully compiled language and come out way ahead in terms of time? No. Did he want to recode to knock off alot of the computing in a very short time using the school's Linux cluster? No. Did he want to take his heaviest subroutines, write them in a compiled language, and call them from Matlab to speed up his calculations dramatically? No. Did he want to know how to rewrite his Matlab code to maximize vectorization and computation efficiency? No.

    He just wanted to throw more power behind what he was comfortable with.

    People will buy this stuff.

  162. Re:Claiming? by level_headed_midwest · · Score: 1

    What wireless card? I popped the Ubuntu 5.10 live CD into a brand-spanking-new Dell laptop (700m) and everything worked perfectly with it from the get-go. I was very surprised!

    --
    Just "gittin-r-done," day after day.
  163. Wow by Anonymous Coward · · Score: 0

    Just the thing to waste thousands of dollars of hardware, countless days, and hundreds of kilowatt hours of power.

  164. I liked the part where ..... by Anonymous Coward · · Score: 0

    ..... the microsoft [microminds] brainless PR Manager demostrated the power of Matlab
    on a Dell Cluster running "Linux"; no doubt that the "Linux" is Red Hat!

    HPC Blue Screen of Death ... Here We Cum! Grab your towles at the door!

    Toodles!

  165. Of Course by rspress · · Score: 1

    Of course most of the power of the supercomputer will be used to run Anti-Virus, Anti-Adware and to download ads to display in the terminal. I can see it now. "Virus detected- Cleaned" "Adware detected- Cleaned" C:|| Run genome project calc "Preparing to run genome project calc but first a word from geritol" "Loading Geritol ad........do you have MSN?............loading..........Try Windows Vista, it's cool..........loading.........."

  166. Oh Stop It! by h2d2 · · Score: 1

    Please STOP refering to the 'Blue Screen of Death', it's long gone and out of fashion. Can't you guys find something new to bash MS about?

    --
    Mozilla stole tabs from NetCaptor. So what? Right?
    1. Re:Oh Stop It! by Legion303 · · Score: 1

      W2K3 still has BSOD issues. I've run into enough of them to know. Getting back on topic, W2K is more stable than 2K3 on purely anecdotal evidence, but the very huge MS company for whom I've maintained both OSs bears it out.

  167. Waah! Lookit meee! Microsoft! Lookit mee! Waaah! by Anonymous Coward · · Score: 0

    Waah! Lookit meee! Don't forget Microsoft! Lookit mee! Waaah!
    Don't look over there.
    Waah! Lookit mee! Don't forget me, Emperor Gates! Waah.

  168. Ah yes, easier to integrate with Linux than Linux by hagbard5235 · · Score: 1

    Since much of HPC today *is* being done on Linux, is he implying that this new version of Windows will be easier to integrate with Linux that Linux?

  169. let me get this straight by Anonymous Coward · · Score: 1, Insightful

    he is comparing windows to linux on super computers?
    so if i want to make a super computer cluster of say 10,000 pcs i immediately have an added cost called the microsoft tax of a minimum of $100. or 10,000,000 usd ..
    am i right?

  170. Re:If you want easy of use pick *ANYTHING* but App by Anonymous Coward · · Score: 0
    Apple's NFS server only supports 64 simultaneous connection.
    Disk I/O still slow as molasses. Grad student discovers the software defaulted to non-buffered disk output (flush to disk).


    I assume you're running on XServes. That's the default in order to ensure the utmost degree of reliability in the case of a power outage. So you've got these levels all built in (not to mention whatever else you can add, like battery backup and a generator):
    • HFS+ journaled filesystem
    • redundant power supplies
    • battery backup for disk controllers so reads/writes and cache can flush before shutdown
    • not even requiring the cache to be flushed by default

    Apple recommends leaving this cache off for reliability and data integrity if something goes awry, but if all you want is speed, then enable it. As for your other problems: if the network filesystem is so important to you for a project like this, I recommend researching it for any potential limitations *before* spending a godzillian dollars on a cluster. When worse comes to worse, though, compile your own NFS, etc. for your own purposes; that's allowed. Remember that you're working on the open Darwin platform.

    I can't speak to the "ease of use" of your Apple installation, but if your only problems were the ones cited above, it's a matter of doing your research so that you can correctly use the tools you've bought.
  171. Re:If you want easy of use pick *ANYTHING* but App by Anonymous Coward · · Score: 0

    We call it an apple tree.
    A mac cluster What the fuck!

    Now thats a cluster fuck.

    DRM

  172. Re:Claiming? by Anonymous Coward · · Score: 0

    I think you are a cluster fuck.

    I am new to linux and I got my sound working.

    I was also new to windows and I got my sound card to work with windows.
    and in the process I learned something.

    be afraid be very afraid.

    gunilla

  173. I like linux but... by Anonymous Coward · · Score: 0
    ... I don't know a single person who is confused by the doubleclick system. And i know alot of really dumb people. to me, it seems like a very natural way to operate, seperating uneditable online content and local objects that may consume an awful lot of resources and provide you with a more significant delay than with just a second click.


    Though i don't agree with stupid shit like hiding the Program Files folder's content by default, i think the doubleclick is an extremely good way to give a buffer zone in case of user error. We're not talking about the inept here, everyone says 'oops' once in a while.

    Doesn't anyone else get kinda bored and click on random spots on the desktop, or is that just me?

    1. Re:I like linux but... by richlv · · Score: 1

      ... I don't know a single person who is confused by the doubleclick system.

      a lot of net-too-savvy computer users i know have at least a couple of times doubleclicked on quicklaunch icons (both windows/kde), links in browser etc.

      i think, i have myself done that too - i just didn't know wether particular linux system was configured for a single or doubleclick. and even though i prefer doubleclick (just used to it ? or is it really better ?), i agree that this causes some sort of cunfusion and definitely is very unconsistent.

      Doesn't anyone else get kinda bored and click on random spots on the desktop, or is that just me?

      it's just you. i almost never see my kde desktop. maybe with kde4 that will change...

      --
      Rich
    2. Re:I like linux but... by Anonymous Coward · · Score: 0
      a lot of net-too-savvy computer users i know have at least a couple of times doubleclicked on quicklaunch icons (both windows/kde), links in browser etc.

      Which might be a problem if it caused those objects to behave any differently. It doesn't.


      it's just you.

      You're serious? no one else just finds themselves playing with the drag-select box? huh... guess that makes me a weirdo.

    3. Re:I like linux but... by richlv · · Score: 1


      Which might be a problem if it caused those objects to behave any differently. It doesn't.


      in some cases you get the application launched twice. if the app is slow already, launching two instances in parallel isn't any faster.
      also, if the aplication locks the profile, second instance might behave differently and confuse the user (for example, mozilla family products)

      You're serious? no one else just finds themselves playing with the drag-select box? huh... guess that makes me a weirdo.

      mhh. as i said, i have no desktop available, so in case i have no desire to do anything more productive i tend to pull out some interesting, but somewhat forgotten problem with, for example, desktop configuration or check wether some project has released a new version. or something.

      --
      Rich
  174. Vacuum tube computers... by Circlotron · · Score: 1

    ...could statistically be expected to fail every half an hour, so I read. Given Microsoft's OS reputation, what's the bet an MS powered supercomputer won't be any better? I mean, we *are* 60 years down the track.

  175. Re:Ah yes, easier to integrate with Linux than Lin by ankarbass · · Score: 1

    Yes, of course, this way you can just run that excel spreadsheet directly on the supercomputer, no need to convert it to a bunch of awkward C code.

    --
    Wanted: Clever sig, top $ paid, all offers considered.
  176. as powerful as linux? by Jump · · Score: 2, Insightful

    "Gates, as always, has high hopes for this new version of Windows, even claiming it to be as powerful and easier to use than Linux." You mean, Windows is becoming even worse now? I thought he usually claimed, Windows is far superiour to just about everything out there.

  177. The title is wrong. by winchester · · Score: 2, Insightful
    Microsoft does not compete in the supercomputer market. Their offering is currently in beta, limited to 128 nodes, and none of the important "traditional" applications used in HPC run on it.

    To give an idea, 128 nodes will give you at most 512 processors (more becomes EXTREMELY inefficient). 512 processors will net you a place between 300 and 500 on the current top 500 list. This will be very different on the list to be released six months for now... such small clusters might not even show up.

    Then there is the user group of HPC systems. It is a VERYsmall market, with a userbase, a group of anministrators and a group of manufacturers traditionally used to UNIX, and now migrating to Linux in droves. Windows is not even duscussed. The announcement of the Windows Compute Cluster edition was cause for great hilarity at the workplace, where jokes like parallel word/excel and high-performance visual basic started floating around. No one will take Microsoft serious in that market.

    Perhaps Microsoft will sell some systems to some manufacturers, like in the automotive or pharmaceutical industry. But these guys already know the ways to traditional vendors selling them Linux clusters, vendors like SGI for instance. CHeck the SGI Manufacturing page.

    So... will Microsoft compete? So far, they announced an operating system for clusters. Important questions remain:

    • Will the OS run headless?
    • What low latency networks will be supported?
    • Are MPI and OpenMP implementations available?
    • what about remote management, remote login and remote copy? (On a side note: why is it that Windows 2003 can't have simple stuff like ssh and scp built in?)
    • what applications will be available?
    I have to wait and see... i don't expect anything substantial to happen... and if Microsoft does this for prestige, they are wasting their money. -
  178. Yes and no. by porkchop_d_clown · · Score: 1

    A cluster is more like an embedded system. You set it up once and forget it for a few years (I should know, I've done it a time or two)

    My experience is that clusters are in the middle of migrating from being a comp-eng thing (where you expect to work at the "bare metal" level) into a more of a pro-sumer thing where a standard UI is comforting even if it burns cycles.

    As for Apple's offerings - I think they illustrate both your attitude and what I describe. My, hrm..., perception is that Apple didn't think there was a point in workgroup clustering either, but more-or-less got dragged into it by their customers.

  179. Re:How much would it cost to be wrong by Viol8 · · Score: 1

    "is the kind of thinking that put IBM where it is today."

    What , you mean the biggest computer corporation on the planet?

  180. As to your other questions by porkchop_d_clown · · Score: 1

    I don't think you understand HPC as much as you think. Remote batch management? Parallel processing? Again, my experience is that the clusters tend to get used for one or two jobs at a time, which is different from a grid. In that case, the MPI process manager and Microsoft's unspecified "management console" are more than sufficient.

    As for interconnects - I know for a fact that gigE (trivial) and Infiniband will be supported (I believe the IB support comes from SilverStorm) and I've heard Myrinet will be supported, but who wants tired old Myrinet when you can have shiny new InfiniBand?

    who, me? work for an IB company? what makes you say that? ;-D

    1. Re:As to your other questions by BigFootApe · · Score: 1
      my experience is that the clusters tend to get used for one or two jobs at a time, which is different from a grid


      This distinction (grid cluster) can become very fuzzy when you have a diverse base of users.

      By remote batch management, I mean

      $ scp /etc/hosts.allow 192.168.0.1:/etc/
      $ ssh 192.168.0.1 killall -HUP inetd

      repeated for 256 machines. This is trivial to do on a Linux box. It is hard (although IANAWE) in Windows. Windows needs a simple and secure means of pushing minor configuration changes out to nodes (you don't want to do major changes in production machines). Running a WSUS server is _not an option_.
  181. The only good thing is that the morons, uh, bosses by crovira · · Score: 1

    are now slowing down and 'evaluating' any 'new' technology.

    They don't understand it but they do understand that the bread line waits for anyone who impacts the bottom line. That means the TCO is being taken into consideration.

    That gives a chance to reject the inappropriate or unwise use of one technology, say using a Beowolf cluster of AMD64Athlon, over another, say using a bunch of slaves on abacuses to calculate the odds for 'intelligent design'.

    For some problems, either approach is inadequate.

    A heavy club with a nail through it though...

    --
    MSBPodcast.com The opinions expressed here are my own. If you don't like 'em... Think up your own stuff.
  182. Easier to use than Linux? by CyberdogOSX · · Score: 0

    Isn't that like saying; "Better smelling than shit"?

  183. Umm exuse me but you appear to be a moran by Anonymous Coward · · Score: 0

    I can compile my code with the INTEL compiles hooked into my Microsoft dev tools with ZERO effort. They all work and play well together.

  184. yes. by porkchop_d_clown · · Score: 1

    I'm assuming (I could be wrong) that this will be part of the management console extensions that MS describes in their press release.

  185. Reporting from SC05 by warrior_s · · Score: 1

    Bill Gates was going to give a live demo at his key-note speech for the new Windows Compute Cluster, but he didnt gave that and the buzz is that he crashed it while pacticing the demo and also crashed the cluster on which it was supposed to run. What a start !!