Become a fan of Slashdot on Facebook

 



Forgot your password?
typodupeerror
Check out the new SourceForge HTML5 internet speed test! No Flash necessary and runs on all devices. ×
Silicon Graphics

SGI to Build Commercial Linux Supercomputers 109

jfinke sent in a link - as did many others - to a brief Yahoo News story about SGI's plans to move into the parallel processing supercomputer marketplace with Linux as their base OS. "We're really pumped up about it," said Beau Vrolyk, senior vice president of SGI's product group. "It represents the beginning of a whole new generation of supercomputer."
This discussion has been archived. No new comments can be posted.

SGI to Build Commercial Linux Supercomputers

Comments Filter:
  • by Signal 11 ( 7608 )
    Had to be said.... BEOWULF! Okay, mark me down now. =)

    --
  • Another great victory for Linux! Linux's growth & world-wide-awareness grows exponentially with each new announcement like this. And it's great to see main-stream, "old style" companies like SGI adopt Linux. Now let's hope that IBM starts shipping only Linux machines w/ Apache web server software...
  • So, the article says that SGI did a demonstration with Itanium chips. Is this true ? I thought that Itanium was only available in simulators ( that's how they're developing Linux/IA-64 ), no silicone yet.

    Anybody care to elaborate ?

  • Actually, it seems that what they're selling is comparable to a Beowulf cluster (if not actually a beowulf cluster). I don't remember the actual quote, but they say that unlike traditional supercomputers, this one is actual a group of 4 computers with Intel Itanium (still sounds funny to say that) processors.

    So, how many companies are demoing computers on this chip? And w/ what OSes? (I know, I'm sure this has been gone over a thousand times, but I wasn't paying attention.)
  • This is great news for the Linux community. However, how can we be sure that SGI wont create a proprietary distribution for its parallell processing systems that may fragment linux?
  • Sorry guys, but I believe SGI needs some mindshare, and that's why they are doing this. Irix scales much better, and so does Solaris and UnixWare. But, granted, they are not as "hot" these days.

    It's a good thing for Linux and SGI, but not because Linux is such a good SMP OS.

    OT: I have just read they are selling Playstation here in Helsinki for 700 FIM (~145 US$) Do you think it's a fair price?


  • by Anonymous Coward
    a) You can't create a proprietary version of the Linux kernel, that would violate the GPL.

    b) There are already proprietary distributions. If you consider that "fragmentation" then SGI's hypothetical distro is a drop in the bucket.

    c) You can fragment the Linux kernel, but if you do so you can't use the word "Linux" to describe it. Linux is a trademark owned by Linus. This would be very bad for marketing. And besides, fragmentation is rarely beneficial (as far as the technology) to either fragment. SGI would be far better off submitting code to Linus, for a host of reasons.
  • Actually, though the comment doesn't add much, it's actually highly relelvant. However, what the heck would you call a cluster of these things as they're already Beowulfs? Megawulf?:)
  • We should see more strength in the PC SMP for linux, *bsd areas. Especially if everything is GPL'ed as I guess it would be...

    ---
  • This has little (if anything) to do with SMP. This is clustering (as in Beowulf). Much easier/cheaper to implement than SMP. Yes, it's hyp, but it's not just hype.
  • Yeah, I was just making reference to the inevitable "oh boy, wouldn't it be great if we could make a beowulf of x?!" AC comments. =)

    --
  • OK, thanks, that part eluded my attention.

    Do you have some bits about the Playstation part?


  • In the UK they go for free[1]. But you do have to buy five games from a selection of not-that-great ones...

    [1] According to an advert I saw in a paper, anyway. Dixons?
  • I forgot to see if I won that spankin' new linux server they were giving away. The drawing was on the 12th.

    But seriously, this gives me a great chance to laugh at my boss. I hope that this means new and better graphics tools for Linux like 3D studio Max and the like. Also, with SGI on board we might start seeing linux support for some of the high end rendering tools, and this will help greatly with linux multimedia development. I am just soooo friggin' happy! I won't have to reboot between tracings of my 3D models now like I do if I use an NT workststion....

    ~Jason Maggard
  • Ah yes, but what about a beowulf cluser of AC's? Let see... a car battery, a bunch of jumper cables, a kiddie pool, and some duct tape might do the job...
  • Now let's hope that IBM starts shipping only Linux machines w/ Apache web server software...

    I think IBM's still making money selling the descendants of that line of computers they came out with in 1963 or so (System/360's descendant, the System/390), as well as the descendants of that line of computers they came out with in the middle or late '70's (System/3x's descendant, the AS/400), so I don't expect that to happen in the near future.

    If customers are willing to pay them money for boxes that aren't Linux machines running Apache, it's not clear that it'd make sense for IBM to refuse to sell them those machines. Perhaps those are all legacy machines, but it may still cost somebody less to buy a non-Linux (or non-UNIX in general) machine than to convert their application right now. (Besides, I wouldn't be surprised to see Linux and Apache - along with Windows and Solaris and HP-UX and Digital^H^H^H^H^H^H^HTru64 UNIX and... and IIS and Netscape Enterprise Server and... - become "legacy systems" some day....)

  • I am a little wary that this may just be a publicity thing. SGI has spun it's Cray division, and wasn't that in part because it wasn't as successful at they had hoped.

    Linux may be full of hype right now, but if they weren't successfully with an already established supercomputing platform (ie. the Cray), how will they manage with something new.

    On the bright side, even if their endevour fails, perhaps we'll get some cool new (and hopefully GPLed) stuff ported over to linux.

    Dana
  • Itanium is available to select few developing companies under NDA, SGI is one of them
  • by Anonymous Coward
    One word: price.

    The whole reason clusters are so popular is that you can build one, for (say) $100,000, with processing power equivalent to a $1,000,000+ Cray.

    Of course, this doesn't apply to everything (see: fine-grained vs. coarsed-grained processing, memory sharing, etc.), but for a large number of applications, clusters are very cost-effective.
  • by Cef ( 28324 ) on Tuesday November 16, 1999 @04:58PM (#1527112)

    A lot of SGI's products (the open source ones) are aimed as toolkits for system management, or are aimed at increasing performance and stability (XFS).

    One such project is a toolkit for obtaining system performance statistics. They write the toolkit, and give it away for free, and then they write the management tool that sits on top of it, and sell that!

    And how better to know just what your cluster of machines is doing, than with monitoring software? And produced by the same company that sold you the hardware? Why not?

    By giving away the toolkit (and it's source), you end up with free improvements, free ports to other platforms, and toolkits that monitor stuff you may never have even thought to monitor in the first place, broadening the scope for your product.

  • by Ami Ganguli ( 921 ) on Tuesday November 16, 1999 @04:58PM (#1527113) Homepage

    Even without SGI, Linux is becoming more scalable with each release. Now SGI is investing real money in the Linux kernel. They're working on projects aimed at implementing the best features of Irix under Linux. It won't happen overnight, but it won't be that long either.

    It makes a whole lot of sense financially (especially for a company that needs to cut costs). They can let the Linux community take care of all the mainstream features that every OS must have, and SGI can focus on the handful of things that are really important to their particular market. Even if SGI ends up being the only maintainer of ccNUMA for Linux, it's still cheaper than having to maintain the entire OS.

    The other exciting thing about Linux is the mainstream applications that are emerging. I know people don't buy an SGI box to run spreadsheets, but it's nice to know that you don't have to keep a Windows PC around just for the odd time you want to run an office productivity application. Linux lets SGI ride the wave of new mainstream developments while still catering to their niche market.

  • I was hoping (after reading carefully) that it's actually not beowulf, but some ingenious SGI soluiton. Beforeyou flame my head off, let me explain: I find it quite difficult to program for beowulf. It's hairy enough to use kernel threads.
    Or is it just me? Is my laziness factor above the average?


  • While this isn't totally on topic, I feel I this is as appropriate a time as any to say this.

    As an owner of a MIPS-based sgi, I don't think sgi is for linux/open source as much as they'd like us to think. I believe this for several reasons-for one, even on their older mips based sgis, they don't seem to want linux(with a working X) to be running on them. Also, beyond Jesse and the kernel dump analyzer tool, nothing has come out of them. Secondly, As several people have pointed out, it really is kind of stupid for sgi to sell linux-based supercomputers(see the other posts to clarify).

    I hate to say it, I think sgi is using Linux as a advertising thing...

    for those who say that this [sgi.com] demonstrates sgi's deep commitment to Linux, think twice...look at the status page [sgi.com], and you'll see that it hasn't been updated in months! So no progress in the MIPS version of Linux.

    just a little rant and raving to make me feel better...I've been really frustrated with my IRIX box the last couple months :-(

  • As far as i can see, fragmentation is not really an issue here. Even if they did create a proprietary distro, wouldn't it only run on other massively parallel supercomputers? Now, i don't know about you, but I would guess that the vast majority of Linux users/developers don't have access to this type of machine, and consequently wouldn't have a use for such a distro.

    Of course, if anyone does happen to have a Cray in their closet, I would most graciously accept a shell account :)
    later,
    hummer
  • This is clearly flamebait -- just because what you say is pro-Linux does not make it anything more than pure FUD. That's right, it's not just for MS anymore, kids.

    Irix is a much more robust and scalable system than Linux. This may change with time, but for the present, it is true; Linux is still lacking in the multi-processor area. And as for Enterprise, well - Irix is proven in the enterprise, whereas Linux is still be looked at as less-than-ready.

    -----------

    "You can't shake the Devil's hand and say you're only kidding."

  • by timothy ( 36799 ) on Tuesday November 16, 1999 @05:22PM (#1527121) Journal
    Ami Ganguli wrote:

    The other exciting thing about Linux is the mainstream applications that are emerging. I know people don't buy an SGI box to run spreadsheets, but it's nice to know that you don't have to keep a Windows PC around just for the odd time you want to run an office productivity application. Linux lets SGI ride the wave of new mainstream developments while still catering to their niche market.


    I think this is an important point.

    When NT started to (self-declaredly) encroach on what had been UNIX workstation territory 3 or 4 years ago, one of the arguments that MS made on its behalf was that people wouldn't need to have a separate box on their desk for common office (read "MS Office") tasks.

    This aspect is one of the things that was hyped in The Analysts' reports (Brown? IDC?) finding lower TCO for NT than for UNIX machines. Now that there are some Web-based application suites semi-available (and more poised to be really available RSN ;)), the idea that anyone would really need a separate box for tapping out a memo is getting sillier anyhow.

    StarOffice, for Free? Heh! Running on a multi-processor SGI sytem? That sounds just about right ...

    timothy

    p.s.(And one day there will be a funny graph of IDC opinions, with an inflection point around this year ...)



  • Please, I'd love to see a Xeon/TNT workstation be used in the high-end graphics design industry. Yeah, it would be fine as a render-engine, but what the hell software would you run on it for design? AC3D? Blender?

    I've got a TNT card in my PIII, running Linux, and while it's a pretty nice setup, it's not a replacement for an SGI system. Just having a Mesa setup won't do it. You need the software and Linux just hasn't got it yet.

    As for the benchmarks: Well, I trust Linux benchmarks by Alan Cox about as much as I trust NT benchmarks by Microsoft.

    -----------

    "You can't shake the Devil's hand and say you're only kidding."

  • >OT: I have just read they are selling Playstation here in Helsinki for 700 FIM (~145 US$) Do you think it's a fair price?

    I don't know what the average price for a PSX is in Europe, but they've been going for 99 dollars here in the U.S. for just about all year.

    Ob:
    It's an operating system!
    It's a cynical ploy to raise stock prices!
    Stop! You're both right! It's Linux!

    Seriously...what is it that Linux does that Irix can't do as well or better?

    --
    "HORSE."
  • >I'll elaborate; it's spelled "silicon" unless you've attended the Dan Quale school of spelling.

    Or if you're talking about breast-based computing...just imagine... a beowulf cluster...

    --
    "HORSE."
  • >Actually, though the comment doesn't add much, it's actually highly relelvant. However, what the heck would you call a cluster of these things as they're already Beowulfs? Megawulf?:)

    Metawulf, duh.
    --
    "HORSE."
  • > We should see more strength in the PC SMP for linux, *bsd areas.

    The article was pretty lean on technical details, but it did say it was talking about a cluster. I would guess they're going the Beowulf route.

    --
    It's October 6th. Where's W2K? Over the horizon again, eh?
  • Linux on MIPS has never been a priority for them. Remember, SGI is shifting all their operations to Linux on Intel (not all at once, but that's the direction they're headed). It doesn't make any sense (for them) to port Linux to MIPS-- they already have IRIX for those machines.

    In any case, don't forget that a lot of people within SGI like Linux a lot, and are very enthusiastic about working professionally with it. Given SGI's creativity-powerhouse culture, it makes perfect sense. I think Linux has a very good friend in SGI.

    Not that that would leave you with much hope for Linux on your machine, but hey-- IRIX's cc is awesome for debugging (with -fullwarn). What other compiler will tell you "variable foo was set but never used?"
  • > I'd take this as evidence Intel gives the you best bang for the buck.

    I solicited some opinions on this recently (we're about to build a Beowulf on a fixed budget), and the (aparently) best unbiased opinion that I got was that Alpha give 2-4 times the performance of Intel, at 2-4 time the price.

    Does anyone concur/demur on that opinion?

    --
    It's October 6th. Where's W2K? Over the horizon again, eh?
  • I'm sick off all the companies that "jump on the Linux bandwagon". Linux isn't a bandwagon, it's an OS, a tool for work and play.

    It seems that when a company hits a dead end, it turns to Linux to liven up the spirit. Do they really have a great interest in Linux? Or are they just trying to save their butts?
  • You write: Irix scales much better, and so does Solaris and UnixWare.

    Whatever systems happen to scale better today, Linux and the free BSDs will scale better than anything else tomorrow, simply because there are 10 zillion developers beavering away at them continuously.

    Furthermore, this continual and rapid improvement will happen regardless of whether or not SGI invest any development time of their own. The economics of this are devastatingly obvious ... devastating to anyone that doesn't jump on the bandwagon that is. Development and post-sales maintenance/support are extremely costly overheads for any manufacturer, so a decision which massively reduces that expense makes great economic sense.

    Furthermore, with Irix behind them, SGI have the pedigree to stand out in the Linux marketplace despite the eventual dilution created by GPL licensing of their enhancements. I doubt if they'll lose many of their old customers, and they are bound to acquire many new ones from their new and greatly expanded Linux audience.

    With this move, I reckon that SGI have their future assured at least until the competition wises up and catches up, and as long as their marketing, pricing and distribution is similarly forward thinking. But that of course remains to be seen.
  • Sounds like an industrial-sized LART. Perfect. =)

    --
  • I cannot say with certainty that this article is about the same machine I'm thinking of, but...
    SGI recently finished a Beowulf for the Ohio Supercomputer Center, and has put it on display at SC99 before actually delivering it to OSC. 128 Xeons for computation (32 SGI 1400L's, each with 4 Xeons, with one more 1400L as an admin node; Myrinet interconnect).
    Since Itanium hasn't actually been released yet, I expect anything anyone debuts in the near future built around Itanium is only debuting 4-color glossies. But, that is the first step before debuting an actual product built around a brand-new processor.
    Christopher A. Bohn
  • Okay, so my spelling sucks tonight. Dont drink and slashdot kids, it only leads to disaster. Hey, we need a misfit mascot for this motto!
  • I am encouraged by this move. SGI put themselves much in my skepticism when they changed from "Silicon Graphics" and started developing NT stations, but have since had another change of leadership. I have been looking for clues that they were back on the right track. I don't know that this is one (a clue, that is), but it lines up with other movements that I like.

    Specifically, the fact that Alias|Wavefront is putting out a Maya render engine for LINUX tells me that graphical tides are turning. An editor cannot be far behind. And I have been very happy with the Alias and SGI combinations I have worked on in the past. If the quality and robustness of graphical interface is there... 'twould be nice!
  • I don't mean to trash Alan. I've got plenty of respect for the guy, considering all he's done for Linux.

    But don't mistake me -- I still don't consider him an objective source for benchmarks. It's nothing personal, but I'd rather see it done by someone without a vested interest.

    -----------

    "You can't shake the Devil's hand and say you're only kidding."

  • by Foogle ( 35117 ) on Tuesday November 16, 1999 @06:11PM (#1527151) Homepage
    Nope, I'm not calling Alan Cox a liar. I'm just a cynic. After all the Mindcraft crap we went through, I figure I'd be a hypocrite to accept benchmarks from anyone except an unbiased source. It's not a personal issue, it's just the way I operate.

    -----------

    "You can't shake the Devil's hand and say you're only kidding."

  • No offense intended here; this comment is meant to be about 25 deg C, and is just a clarification for what seems to be a common misconception.

    No application can make use of a parallel computer without a programmer first identifying the inherent concurrency. There are pre-processors that attempt to do this for you, such as BERT 77 [plogic.com], but IMO that cannot replace the intuition and basic understanding of the problem that a human programmer provides (with all due respect to Doug & everyone else at Paralogic). Once that concurrency is identified, then implementing it using the MPI or PVM API, or using HPF directives, or using pthreads (depending on your platform) should be fairly straight-forward.

    Computers are not magic! You cannot feed a program written for a sequential computer into a parallel computer and expect speedup, period. At run-time, the computer just doesn't have sufficient view of the application to even attempt to identify the concurrency for you, and it certainly doesn't have the cycles to spare to decide what aspects of the concurrency should or shouldn't be parallelized. The best you can do if you won't or can't parallelize your code is to submit multiple instances of the application, with different inputs, to the batch scheduler and obtain a greater throughput.


    Christopher A. Bohn
  • In an interview over at the ZDTV booth today, Linus was asked what the hardest task that he is planning on adding into the Kernel. He stated NUMA (Non-Uniform memory access), and who uses ccNUMA? SGI. [sgi.com]

    I've got the feeling that SGI and Linus are well in communication.
  • I find this interesting, mainly because it makes me think of previous work SGI has done in the large-computer arena; more to the point, I'm wondering if they're turning to this approach in part because some of their previous work has suffered deeply from a lack of linear scalability.

    I'm thinking in particular of the Blue Mountain machine at LANL -- this was/is a big set of SGI Origin 2000 boxes; I don't actually remember offhand, but I seem to recall that they ran Irix, etc, and used typical MPI message-passing crap for most applications. Never mind. Anyway, point is that there were some problems with getting a linear performance increase on the machine as more and more processors were added; we're talking, incidentally, of thousands of processors. Some problems were to be expected -- clearly it's naive to expect totally linear speedup as you increase the number of processors, even for extremely well parallelized code. But some of the things we saw were just ridiculous -- there was, for instance, a sort of catastrophic breakdown at about 4000 processors; this has probably been solved for some time now, but not without an awful lot of work being done to do so. For some more info about this, check out this page at LANL [lanl.gov], among others.

    Don't get the impression, btw, that I'm against SGI getting into this area -- I happen to like the company quite a bit, think they have some extraordinary products, etc etc. I'm just a little wary these days when I see them talking about massively parallel machines.

    Cheers.

  • Part of being able to scale, IMO, is to be able to handle a great number of threads. This is very important to the type of research I do (complex systems simulations, where I can have 1000+ threads).

    Linux is at a major disadvantage here... it only supports course grained/one-to-one thread mapping (one thread to one process). More scalable systems, like Irix and Solaris, use a many-to-many mapping (groups of userland threads mapped to multiple kernel threads/processes).

    Is there any plans to correct this deficiency?

    --

    "A society that will trade a little liberty for a little order will deserve neither and lose both."

  • There aren't going to be 10 zillion developers working on machines with hundreds of processors, multiple terabytes of disk, and who-knows-how-much tape storage. Your average developer isn't going to have access to such a machine.

    It may be true that Linux and the BSDs will scale to this level of system in the future, but it will be due to companies like SGI spending lots of money funding that development work.

    Personally, I don't see how SGI's new-found Linux enthusiasm is evidence of an economically sound strategy. Re-implementing big iron features of IRIX for for Linux is going to be costly and time-consuming.

    But, I think that SGI is after the applications that will run on Linux, and believes that it's worth it to fix up Linux to run on big machines in order to get access to that application base.
  • Last night the Vancouver Linux users group presented a talk by Dr. John Mashey from SGI. It was a very interesting talk, and convinced me that SGI does "get it" and is truely dedicated to Linux and the open-source concept.

    I don't think that MIPS/Linux is a priority for SGI. They see MIPS/Irix at the 'top' of the market, and Intel(x86 and itanic)/Linux at the 'bottom' but moving up over the next few years. They know it's coming, and they'd rather spend resources to influence its development than be a passive by-stander as they are with NT. So, they feed features and ideas from IRIX into Linux, while continuing to sell and maintain the high-end IRIX systems as long as there's still a market (which consists more of corporate uses like automotive crash-test simulation than of TV and movies).

    He also said that several of SGI's customers want Linux, and want certain features in it (like raw I/O for big databases). So, SGI either helps develop this and keeps the customers, or loses them.

    The XFS filesystem is one of their major contributions, and he said that one of the main delays there is that they are going through the code to remove any sections that could cause patent headaches in the future. One hopes the result will be slightly more usable than the initial Mozilla...

    ---
    disclaimer: these were my impressions of Dr. Mashey's talk; they may or may not actually represent his views, which may or may not be those of SGI.
  • That's right, they aren't in silicone yet. The Itaniums will be available in silicone towards the end of next year... I hear that Pamela Anderson will be using them to power her machine gun jubblies in Barb Wire II.

    "... message passing as the fundamental operation of the OS is just an excercise in computer science masturbation."

  • I can't recall whether the Y-MP used floroinert, I think it was the standard freon way of cooling. If it does use floroinert, you should be able to get it from 3M, albeit it won't be in the aisle with the scotch tape. Come to think of it, I am fairly certain that it doesn't use floroinert because the Y-MP doesn't have the waterfall stuff that the Cray 2 and machines after the Y-MP (C-90,T-90, etc) have.

    Did you get the motor generators with the machine? They are big (and *VERY* *VERY* loud) generators that generate the 400hz power that the Cray's use. They are always kept in a nearby room with serious sound insulation.
  • as another nail in SGI's coffin? This will get them some attention but unfortunately not get them much business. I don't want Linux on a super computer, it would cease to truely be linux if you changed it enough to actually work as a super computer. What makes a computer super is not it's total operations in a second but how efficiently it handles those operations. Crays use(d) vector systems, the chips are mostly SIMD circuitry which means you can get more usable data from a single instruction. Until recently PC class chips had never touched SIMD, they were all SISD circuitry. This means the chips in Crays were the opposite of what a PIII is today, it was mostly SIMD circuitry with small SISD units (where the PIII is mostly SISD with one or two SIMD units). Besides the actual hardware difference, Linux just does not have the super computer mentality. Irix was designed to scale on hundreds of processors and address hundreds of gigs of RAM, Linux makes poor use (comparitively) of only hundreds of megs of RAM. If Linux were tweaked by SGI's old Irix people (do they even work there anymore?) it would really cease to be Linux. You have to remember that Linux was originally designed to be a flavour of unix that ran on PC architecture, not power hungry-underground complex requiring mainframes. This is no kind of slur against Linux it is focused on SGI. It's a shame they have been reduced to trying to get under someone else's PR umbrella to make headlines. If someone offered me the choice of an old Origin 2000 or one of their Linux super comps I would go for the Origin based almost entirely on principle.
  • Er... As an owner of a MIPS-based SGI too, I need Linux on it like I need a hole in the head. I'm happy to see SGI support Linux on the Intel platform, but I don't particularly look forward to or care to see Linux on MIPS, for whatever long MIPS has left, which may not be that much.

    I understand that maybe IRIX isn't the best OS around from a security standpoint, but as a workstation OS for a single user, is absolutely peerless, with only BeOS and the AmigaOS coming close (although I have no first hand experience of Solaris, I'd be happy to add it to the list).
  • Sun, SCO/IBM (Monterey), Mickeysoft (rumors have it that itactually doesn't boot at all), HP ?


  • Why do you need this in cc whats wrong with using lint:

    braveheart.UK$ cc -Xc notused.c -o notused
    braveheart.UK$ lint notused.c

    argument unused in function
    (3) argc in main
    (3) argv in main

    set but not used in function
    (5) foo in main
    braveheart.UK$ cat notused.c
    #include

    int main(int argc, char** argv)
    {
    int foo = 1;

    return 0;
    }
  • The fact that SGI is planning linux based supercomputer clusters is old news. It's been on SGI's site for months. Whats new and exciting here is that they have actually demonstrated the first IA64 technical application running and the first cluster on IA64 and done it using 64 bit Linux/Beowulf.

    Two big firsts on IA64 for Linux and I think it's a great indication that Linux will be the first released OS for IA64.

    Some more info:

    sgi's announcement [sgi.com]

    Project Trillian [ia64linux.com] which is porting the linux kernel to IA64.

  • by Anonymous Coward
    But Linus didn't see Linux would be ported to non-intel architecture when he released it. Now Linux runs on so many platforms and is beginning to get into embeded arena, I don't see why SGI adding/modifying Linux to make it runs on supercomputer/SIMD etc a problem. The code can be made even more moduler to accomodate changes required for such tasks.

    I think from SGI's view point is that they want to focus on the scaling department and leave the rest (common code like device drivers etc) to the Linux community. Eventually it should be a win-win situation when Linus and co see their changes worthy to be merged into main tree. Who knows, may be they can help improve Linux SMP performance on the way. They're the experts, isn't it?

  • Note that what you are proposing is not all that simple: _increasing modularity_ and reworking the _SMP_ scalability of the kernel invovles _major changes_ to the kernel structure. Here lies a great risk - that the changes will be so sweeping, that the linux kernel structure will fragment: we'll have the "Linus" kernel, and the "SGI" kernel. Hopefully if this happens, the changes will merge into Linus's kernel quickly, otherwise if SGI release their changes to quickly we'll have irreversible fragmentation.
  • by EngrBohn ( 5364 )
    Boy my foot sure tastes good!
    Just took a look at the original press release ... demonstrating, Itanium, IA-64, NCSA ... looks like
    • This is not the OSC Beowulf that they're also showing off
    • They did get their hands on Itania (Itaniums?) and are demonstrating something solid, with mass, and displacing real air.

    Christopher A. Bohn
  • now, more that ever, the new logo applies.
  • by Molly ( 32733 )
    Wouldn't a cluster of Beowulfs still be a Beowulf, just with a more exciting network topology?

    Molly.
  • Can't resist this though, how many SGI O2ks are in the top500.org list compared to intel boxes...

    One last thing, I've got enough balls to post with my identity; only scared little liars are afraid of standing behind what they say.
  • I feel that the release / announcement of _new_ SGI Linux boxes is a good thing because it was something planned some time ago; thus, even though they've had a change of leadership, they're still committed to this as a path. Personally, I think they've made a good decision: "we're a UNIX company and we're good at it, we'd never win by trying to fight the NT fight."

    My view on the SGI Linux effort on Intel is that they're trying to do exactly what's been mentioned here: grab mindshare (and thereby marketshare).

    This is the image that they want to put in your mind: a spectrum of machines, rangine from your desktop (you're running Linux on your desktop, right? ;-), through file & compute servers and graphics workstations all the way up to monster multi-processor visualization and GP computing beasts. The interesting thing is that some of those machines would be running Linux on Intel, some IRIX on MIPS - they want those lines to blur. Even better, the machines could be source level compatible - not quite as sweet as a MIPS-only shop (just drop the binaries anywhere), but just a recompile away.... ;-)

    Finally, I don't think SGI is spending money on Linux on MIPS because they understand the Linux community: if the Linux community at large really really wanted Linux on MIPS, they'd make it. If there were a loud movement that said "yes, we do want it but we need more information to make it sweet", I could see them coming across. Open Source = good business for them.

    Finally finally ;-), I'd bet they'd dearly love to see their Intel-based Visual Workstations running Linux with full hardware-supported graphics (last I checked, it could only do software-based). Any takers...?
  • A whole bunch of Y-MPs used florinert. For this guy's sake, I hope he got a Y-MP EL. That is the only one that he could hope to get running in his garage without spending a fortune on infrastructure. The EL was labeled a "departmental supercomputer". You could plug it into a 220V socket and it had enough fans to cool the system boards. For you Crayons out there, you may remember the marketing department's nickname for the Y-MP EL - "dime bag". Heh.
  • The message SGI is giving the market is indeed that IRIX may be sufficient now, but with a bit of effort Linux wiull probably surpass it. Anyone who'd been considering buying SGI will now go to someone who's 100% behind their systems (and I'm sure there will be many current customers who'll shift because of this as well). Idiots.
  • The message SGI is giving the market is indeed that IRIX may be sufficient now, but
    with a bit of effort Linux wiull probably surpass it. Anyone who'd been considering
    buying SGI will now go to someone who's 100% behind their systems (and I'm sure
    there will be many current customers who'll shift because of this as well). Idiots.


    I don't necessarily think so. For one, "a bit of effort" is prolly going to turn out to be a year or two, and most people buying workstations will need their solutions solved now.
  • There aren't going to be 10 zillion developers working on machines with hundreds of processors, multiple terabytes of disk, and who-knows-how-much tape storage. Your average developer isn't going to have access to such a machine.

    The resources available to the average developer are irrelevant. What matters is that just one good developer has them available, and the lesson we've learned from Beowulf is that not just one but a lot of developers will find themselves suitably resourced.

Programming is an unnatural act.

Working...