Slashdot Mirror

It has amongst educated consumers i.e. server admins, researchers, and the top500.
Also, the Android platform is based on Java running on Linux. Where were the wince phones on this list again?

So where's that 8P Intel PC? Granted, it's less of an issue now that there are 6 core CPUs out there, but let's not forget that Intel still can't effectively put one together.

Um... what planet are you from? On mine, Intel has been shipping Nehalem-EX for a while now. 8 cores on one die, with glueless support for up to 8 CPU sockets on a board.

Let's look at that Top 500 list. The top 2 are AMD systems, and that the 3 Intel systems are in the bottom 5.

ROFL. You conveniently didn't link to http://www.top500.org/charts/list/34/procfam , which shows that Intel thoroughly dominates the Top 500 supercomputer list. There's a lot more than just 3 Intel based computers in the list!

Yes, the top 2 happen to be AMD; it's more or less meaningless. Massively parallel supercomputers are more about the interconnect tying all the processors together than the processors themselves. Did you notice that a ton of the machines in the Top500 are Blue Gene? IBM Blue Gene supercomputers are one of the few left that use a custom designed processor instead of a commodity CPU. That custom CPU isn't anything special, though: it's a PowerPC core derived not from IBM's firebreathing POWERn line, but instead from their old PPC440 low power embedded control CPU. You would not want to use one of the CPU nodes from a Blue Gene to run even a web browser. But Blue Gene gangs metric fucktons of them together and the result is a competitive supercomputer.

The top 2 machines in the list are at the top because the organizations which commissioned them (United States DOE and Oak Ridge National Laboratory) have very large budgets to buy larger supercomputers than anyone else, so they can buy more nodes. Look at the processor counts on each.

(Also note that #2 isn't a pure Opteron machine, it's an Opteron/PowerCell hybrid. I'd guess the Opterons are for communications and the Cell processors are what provide the FP grunt.)

If I want to browse the web and not heat my house at the same time, AMD really does offer the better chip, and cheaper by far too.

Once again, what planet are you on? On mine, for the last several years, virtually every review comparing an AMD notebook to an Intel notebook has found that roughly equivalent Intel notebooks outperform AMD's and also have much better battery life. You need to pull your head out of the sand, it isn't the P4 vs. Athlon days any more.

If I want to play that FPS and have an extra frame or two, then the Intel chip is a winner. If I want to participate in something like Folding@Home, then Intel is a winner if power and heat aren't considered. If I'm rendering video, Intel might be a winner, depends on whether heat matters.

Available evidence suggests that Intel is far, far ahead in terms of power efficiency at full load. In other words, if you do Folding @ Home, and divide performance by the watts used, Intel has a much better perf/W ratio than AMD. Since AMD and Intel ship similar wattage rating CPUs, that translates into an absolute performance advantage for Intel too.

Once again, Intel has stuff just a bit better than the P4 now. The memes you blindly parrot are a bit out of date.

So where's that 8P Intel PC? Granted, it's less of an issue now that there are 6 core CPUs out there, but let's not forget that Intel still can't effectively put one together.

Let's look at that Top 500 list. The top 2 are AMD systems, and that the 3 Intel systems are in the bottom 5.

If I want to browse the web and not heat my house at the same time, AMD really does offer the better chip, and cheaper by far too.

If I want to play that FPS and have an extra frame or two, then the Intel chip is a winner. If I want to participate in something like Folding@Home, then Intel is a winner if power and heat aren't considered. If I'm rendering video, Intel might be a winner, depends on whether heat matters.

Very few things even most on /. would utilize a computer for will only see an Intel advantage maybe 1% of the time.

After all - does using an AMD or Intel chip make any difference rendering /.?

Are you considering graphics cards as gaming accessories or graphics cards as parallel math coprocessors for the medium-end number crunching on a budget market? If you consider the latter then drivers for graphics card, which bring support for OpenCL, will make linux a worthwhile market segment. Where do you find people crunching numbers? Windows? OSX? No. All the cool kids crunch numbers with linux.

... for the multi-core era.

Too bad it is not microsoft: http://www.top500.org/stats/list/34/osfam

Er ... FAIL

Top 500 stats show that Intel has over 80% of the top 500.

Yeah, the top 3 use Opterons, although one apparently is Cell-based and just uses relatively slow Opterons for GP nodes.

Intel dominates in smaller supercomputers, but the ones that put the most cores in one spot use AMD. Most of the others in the top 10 are BlueGene (PowerPC), and Intel rounds out 3 out of the top 10, although one of those machines relies more on Radeons to do the heavy lifting.

Of course, most top supercomputers take years to plan and build, so we'll probably see Intel creeping back up there as the years go by, unless AMD can recapture a favorable price/performance/power ratio. AMD's real secret in doing so well in the HPC market has been their good NUMA capability, lower power consumption in active work loads (Intel is generally better at idling), and lower per unit cost.

Er ... FAIL

Top 500 stats show that Intel has over 80% of the top 500.

Dunno what _you_ mean by high-end computers, but of the top 500 super computers, 446 run some flavour of linux, only 5 run a Microsoft OS.

Reference:
http://www.top500.org/stats/list/34/osfam

Actually, you're wrong. Not all from the Top100 clusters are running Linux.

There's at least one running Windows. Taking HPC as a whole, there's also AIX, Super-UX, OpenSolaris (reference: http://www.top500.org/stats/list/34/os).

Actually, you're wrong. Not all from the Top100 clusters are running Linux.

There's at least one running Windows. Taking HPC as a whole, there's also AIX, Super-UX, OpenSolaris (reference: http://www.top500.org/stats/list/34/os).

The US isn't the world's superpower?

Name another state that can project tens to hundreds of thousands of troops across the planet and fight for eight years.
Name another state that has more than three aircraft carriers.
Name another state that has more than ten aircraft carrier battlegroups.
Name another state with more than half of the top 500 super computers - http://www.top500.org/stats/list/34/countries

The United States has a list of strengths no other nation or union of nations possesses. Russia has the natural resources, military technology and nukes but not the industrial base and ability to project power. While the US got involved with Iraq, Serbia, Afghanistan while bolstering South Korea, Israel and Kuwait, Russia was bogged down in Chechnya.

The EU has the industrial might and military technology and a good number of nukes, but little ability to project power and no political will to do so. Only the UK and France regularly use offensive military operations, but their militaries are a fraction of the US. The UK has maybe 3 division equivalents while France has 2.

China has older industrial might, older military technology and some nukes, but like Russia and the EU and everyone else can't project power. Going across the Straights of Taiwan will be the biggest thing China could do and even in the next 20 years, thats iffy.

CRAY is hardly tarnished

http://www.top500.org/system/10186 The machine quoted in TFA is quoting single precision. Currently the ATI boards trounce the Nvidia boards in double precision. The next GPU cluster down the list is Nvidia based at #56 http://www.top500.org/site/690

http://www.top500.org/system/10186 The machine quoted in TFA is quoting single precision. Currently the ATI boards trounce the Nvidia boards in double precision. The next GPU cluster down the list is Nvidia based at #56 http://www.top500.org/site/690

Ummm isn't this just a ridiculously powerful desktop computer rather than a super computer? The current 500th super computer on the top500 list is this machine which has a Rmax of 17 Tflops and an Rpeak of just over 37.6. Now its impressive that this desktop system has 1/37th of the power of the lowest machine on the super computer list... but does that really make it a super computer? Moore's Law says that it will take around 10 years for this desktop box to evolve to the power of that current bottom top500 box. So in other words its 10 years behind the performance of the current 500th best super computer.

If its because it hits 1 Tflops then in a few years time you'll have mobile phone "super computers" as Moore's Law is still moving onwards.

This is a very very fast desktop computer suited to certain simulation elements which are GPU intensive. Nice box, fast box.... but not a real modern super computer.

I seem to be having Deja'vu with a little company called SiCortex that no longer exists (the Wikipedia article is out of date). Why? Because nobody wanted to rewrite their software for a machine less than 0.1% of the market used. The other reason is that most software in the HPC world is written to be GCC/Intel compatible so porting to PGI was interesting. This chip might go somewhere if they can market it as a coprocessor with BLAS libraries, however even if they try to do that their going up against IBM with its CELL blades and it's mixed platform BLAS as used by Roadrunner which is currently the number one on the top500.

The DOE and DARPA (and others) are huge users of HPC (high performance computing) applications. The have a vested interest in having the state of the art advance in parallel computing and so they tend to provide lots of research grants to fund that. They also routinely let outsiders use some of their computing facilities for the same reason (not all of their labs do classified work). There are many computing facilities that need enormous computing power as shown on the Top 500 list. But they are seeing that there are times where researchers need computational power, but not at such a large scale and not for long periods of time. If medium powered computational facilities could be made available to researchers cheaply and quickly, they would be widely used.

Not sure why you would say that unless you aren't aware that PS3s are being used to make HPC clusters for scientific computing:
http://www.top500.org/blog/2007/10/27/rough_guide_scientific_computing_playstation_3

Sure, you could do it with a cluster of workstations. You would need some insane interconnects. OR, you could just buy this pre-configured system from SuperMicro with dual quad-core Nehalems and 4 Nvidia Tesla C1060 GPU Cards. That's 960 thread processors @1.3 GHz if you don't overclock, 16GB of DDR3 @ 1.6 GHz on a 512 bit bus, 16 threads of system CPU with up to 96GB of system RAM. It pulls close to 4 TFLOPS, in a desktop machine. You probably could break into the top500 with ten of them with decent interconnects since the #500 spot is Rmax 17.09 TFLOPS and Rpeak 37.64 TFLOPS. If you prefer a top 3 OEM, you can get that in a Z800 workstation from HP.

To put that in a time scale for you, that one desktop available today by itself would have easily been one of the top 100 supercomputers in the world only five years ago and would still have been in the top500 3 and a half years ago.

A little spendy for a wordprocessing and light spreadsheets, but a sweet piece of gear nonetheless.

Sure, you could do it with a cluster of workstations. You would need some insane interconnects. OR, you could just buy this pre-configured system from SuperMicro with dual quad-core Nehalems and 4 Nvidia Tesla C1060 GPU Cards. That's 960 thread processors @1.3 GHz if you don't overclock, 16GB of DDR3 @ 1.6 GHz on a 512 bit bus, 16 threads of system CPU with up to 96GB of system RAM. It pulls close to 4 TFLOPS, in a desktop machine. You probably could break into the top500 with ten of them with decent interconnects since the #500 spot is Rmax 17.09 TFLOPS and Rpeak 37.64 TFLOPS. If you prefer a top 3 OEM, you can get that in a Z800 workstation from HP.

To put that in a time scale for you, that one desktop available today by itself would have easily been one of the top 100 supercomputers in the world only five years ago and would still have been in the top500 3 and a half years ago.

A little spendy for a wordprocessing and light spreadsheets, but a sweet piece of gear nonetheless.

Sure, you could do it with a cluster of workstations. You would need some insane interconnects. OR, you could just buy this pre-configured system from SuperMicro with dual quad-core Nehalems and 4 Nvidia Tesla C1060 GPU Cards. That's 960 thread processors @1.3 GHz if you don't overclock, 16GB of DDR3 @ 1.6 GHz on a 512 bit bus, 16 threads of system CPU with up to 96GB of system RAM. It pulls close to 4 TFLOPS, in a desktop machine. You probably could break into the top500 with ten of them with decent interconnects since the #500 spot is Rmax 17.09 TFLOPS and Rpeak 37.64 TFLOPS. If you prefer a top 3 OEM, you can get that in a Z800 workstation from HP.

To put that in a time scale for you, that one desktop available today by itself would have easily been one of the top 100 supercomputers in the world only five years ago and would still have been in the top500 3 and a half years ago.

A little spendy for a wordprocessing and light spreadsheets, but a sweet piece of gear nonetheless.

Small systems with limited hardware as in more than 88% of the current top 500?
http://www.top500.org/stats/list/33/osfam
Perhaps you should consider moving from your planet to the real world.

1 Exaflop matches up with some estimates of the net raw processing power of the human brain. Note I say raw and net, because that excludes likely optimization methods and weeding out parts of the mind that don't need to be simulated 1:1 (retina for example).

1 Exabyte far exceeds the memory capacity (by 1000x) to store a human brain... uncompressed.

They plan to have this thing running in 10 years.

$1000 computing equipment is hovering around 5-6 years behind Top500 http://www.top500.org/ computers (maybe less these days when you consider a single $100 graphics card bests 1997's 1 teraflop supercomputer).

6-7 years later (at a guess), when all the Top500 supercomputers reach at least the level of 1exaflop+1exabyte, their total storage capacity could store 1 billion human minds.

All I know now is my brain hurts.

The fact is that government did move to prop up many mainframe makers, and even more so with the makers of supercomputers which long ago displaced mainframes as the largest and most expensive systems. It's still happening today. Go look at the Top 500 lists, and you'll see that practically all of the top systems are government-owned. Thinking Machines would never have gotten off the ground without extensive government support, Cray/SGI wouldn't have survived the 90s, and let's not forget DARPA's contributions. Government has contributed positively to innovation in computing, not caused it to stagnate. If the government had shown any inclination to get involved in the auto industry the way they have been involved in computing, we'd all be driving all-electric or hydrogen-fueled cars today, supported by an appropriate recharging/refueling infrastructure and complemented by a robust cargo/mass-transit infrastructure. Grove's an idiot.

It may be that the software for turning a pile of Linux boxes into a rendering farm is free or less expensive or more efficient than the equivalent for Windows.

Indeed. It's not a coincidence that only 5 of Top500's list are pure Windows environments.

You may not be aware of it but Linux outperforms Windows on any platform, 88.6% of the top 500 computers in the world run Linux (June 2009).

FYI the simulation is running on this machine: http://top500.org/system/7388

http://www.top500.org/charts/list/33/osfam

Somebody better tell the admins of the computers on this list they're wasting their time.

You'd think that somebody that's light-years ahead when it comes to parallel processing would rule the roost in the Top 500 supercomputer list. I'm sure there's a good explanation, though....just waiting to hear it. :)

They're in a bind no matter what way you look at it. They've saturated their market three times over. There's no room left for growth in the places where people have money to pay for a desktop OS, and all the people in the other places have tried a pirated Vista already. In the supercompute market their share is 1% despite coming out with their own supercomputer OS(*), and in the server room they're not holding their own either. Their traditional hardware and software partners are starting to come out with their own branded Linux distributions. Because of the Sendo thing they're getting nowhere on the phone.

If Vista 7 tanks, they're in a world of hurt. Like a wise man once said... Outlook not so good.

(*)Some people say that Windows' place on this list is mostly a result of marketing, where the supercomputer sites were given some subsidy to build their supercomputer, with the caveat that they had to report to the Top500 with "Can Run" Windows HPC, and with the Windows HPC benchmark. But for serious work of course they run Linux.

They're in a bind no matter what way you look at it. They've saturated their market three times over. There's no room left for growth in the places where people have money to pay for a desktop OS, and all the people in the other places have tried a pirated Vista already. In the supercompute market their share is 1% despite coming out with their own supercomputer OS(*), and in the server room they're not holding their own either. Their traditional hardware and software partners are starting to come out with their own branded Linux distributions. Because of the Sendo thing they're getting nowhere on the phone.

If Vista 7 tanks, they're in a world of hurt. Like a wise man once said... Outlook not so good.

(*)Some people say that Windows' place on this list is mostly a result of marketing, where the supercomputer sites were given some subsidy to build their supercomputer, with the caveat that they had to report to the Top500 with "Can Run" Windows HPC, and with the Windows HPC benchmark. But for serious work of course they run Linux.

SGI and Cray (and for that matter Sun and IBM) run Linux on some of the most powerful computers in the world. In fact, the in the November 2008 top 500 supercomputers list 9 of the top 10 run Linux (CNL is Compute Node Linux, a light kernel used for cluster nodes). The tenth runs... well MS Windows HPC... Go figure. There's actually a Sun machine in that top 10 list. It runs Linux too.

Never underestimate the bandwidth of a station wagon full of backup tapes. Or, in modern terms, NAS. Nothing replaces the obvious answer: Lend your NAS to a friend with equal quality standards for a few days, and let nature take its course. If that friend doesn't have your rip, back up the NAS to your media center and repeat until all your friends have offsite backups of each other's valuable media for disaster recovery purposes and you have what you want.

Not that I would advocate watching the backups of DVD's that you don't have purchased media for... oh, no. That would be immoral. Not quite as immoral as charging $16 bucks for Waterworld: Director's Cut or Glitter [4:3], but immoral still.

After all these years why are we even having this talk? Kids these days. /lawn, onion, etc.

Since I'm already offtopic, I might as well go whole hog: using an old computer, the free OpenFiler app and a 4 port ESATA card you can turn 4 of these NAS+eSATA devices into RAIN: a Redundant Array of Independent NAS, for 27TB of massively parallel striped redundant hot swappable SATA goodness for about 6 grand served up as iSCSI. If you've been pricing that class of storage in the enterprise lately, you should now be going... wait, what did he say? Yes, I did say iSCSI SAN for $250/TB with redundancy, failover, striped performance and a scalable architecture that goes as big as you like. Yes it includes web management, clustering, unlimited snapshots, resizeable LUNs, static and dynamic replication and all the other SAN goodness. Though I'm not quite sure about data dedupe yet, at this rate for raw storage does it matter? Data dedupe is about making the most of that precious investment. If the investment is out of petty cash you don't need that kludge any more than you need Full Disk Compression, RLE, or any of those technologies built to get around the high cost of storage. Of course it runs in Linux, and of course it's available as a virtual machine, and naturally since the software is FREE there's no per-terabyte licensing. The license doesn't expire, run out, require keys or maintenance or accounting, it doesn't require a license server and when the array needs expansion or an upgrade in technology they can't tell you that you have to upgrade to the new version and buy licenses all over again. It's all about knowing what you're doing - a lot of the Top500 use OpenFiler to serve disk to their supercomputers.

Sorry, but you are coming across as a dick.

1) Who cares if it is 'certified unix', all this means is that it "[conforms] to the SUSv3 and POSIX 1003.1 specifications for the C API, Shell Utilities, and Threads". Guess what, so does HPUX. As a unfortunate user of HPUX it is hard to describe how bad that OS is. So, this certification stamp in no way means this is a good, reliable, or stable OS.

2) Never tried

3) Yes, you can do things with a OSX you can't do on linux/windows. Surprise! You can do things on linux and windows server which you can't do on OSX. I don't see the point here...? BTW, xgrid is not the only distributed computing solution. See http://www.top500.org/ for more info.

4) As a gentoo user, I totally agree. emerge -u -e world has screwed me so often I feel like we are married.

linux has huge advantages on the high end, actually e.g. its O(1) scheduler is a simple but amazing win.

Microsoft are busily trying to buy/bribe themselves into the HPC world, but they have a lot of technical catching up to do, there's a reason nearly all of the world's most powerful/huge computer systems now run linux.

http://www.top500.org/charts/list/32/osfam

If you have a look at http://top500.org/lists/2008/11/performance_development it takes more than 6 years to get 10 times actual performance (quicker than Moore's law, hrm). Given that the actual top is at 1PF, going to 20 in 3.5 years is quite an achievement.

What I kind of read between the lines from your post, is that nowadays it is considered normal that everyone tries to use MS Windows for all tasks, and that the idea that certain programs run on other OSes/computer systems is just not a concept people encounter when they start to study.

I can't imagine that in the Netherlands scientists get no exposure whatsoever with various and sundry ancient or modern OSes besides MS Windows. But as I said, maybe my bias shows because I'm a bit of a Linux fanatic :-)

I remember when there was talk of MS Windows computers being used to run Gaussian (a program used for large computations on molecules), and IIRC the program had to be painstakingly "dumbed down" to 32-bit file I/O because MS Windows couldn't deal with files larger than 4Gb. This was decades ago, I'm sure it must have improved in the mean time, but still.

What I really mean is: you should use the computer systems for your work that get your work done; and the idea that all possible kinds of scientific work can be shoehorned to be done on off-the-shelf MS Windows systems, is ridiculous to me; it doesn't even run on more than one CPU architecture AFAIK, and I'm not convinced it can run a calculation for a month without crashing (I'm told MS Windows XP is much much better nowadays so I could be wrong on this). Look at the top-500 supercomputers sorted by OS family and you'll see what I mean.

Hmm. all that said, I AM impressed that #10 on that list is a supercomputer that runs MS Windows. Would be nice to see a performance comparison between some protein folding work on top of GNU/Linux and on top of MS Windows super-server-version, on that Dawning 5000a cluster

What I kind of read between the lines from your post, is that nowadays it is considered normal that everyone tries to use MS Windows for all tasks, and that the idea that certain programs run on other OSes/computer systems is just not a concept people encounter when they start to study.

I can't imagine that in the Netherlands scientists get no exposure whatsoever with various and sundry ancient or modern OSes besides MS Windows. But as I said, maybe my bias shows because I'm a bit of a Linux fanatic :-)

I remember when there was talk of MS Windows computers being used to run Gaussian (a program used for large computations on molecules), and IIRC the program had to be painstakingly "dumbed down" to 32-bit file I/O because MS Windows couldn't deal with files larger than 4Gb. This was decades ago, I'm sure it must have improved in the mean time, but still.

What I really mean is: you should use the computer systems for your work that get your work done; and the idea that all possible kinds of scientific work can be shoehorned to be done on off-the-shelf MS Windows systems, is ridiculous to me; it doesn't even run on more than one CPU architecture AFAIK, and I'm not convinced it can run a calculation for a month without crashing (I'm told MS Windows XP is much much better nowadays so I could be wrong on this). Look at the top-500 supercomputers sorted by OS family and you'll see what I mean.

Hmm. all that said, I AM impressed that #10 on that list is a supercomputer that runs MS Windows. Would be nice to see a performance comparison between some protein folding work on top of GNU/Linux and on top of MS Windows super-server-version, on that Dawning 5000a cluster

That was not a troll. At worst it was a funny. Has /. moderation been taken over by Redmond?

Linux share of the Top500 is 87.8%. Linux share keeps growing.

Somebody is trying to use the /. moderation system to hide the truth. Please mod them down.

Wikipedia claims that a supercomputer "is a computer at the forefront current processing capability" http://en.wikipedia.org/wiki/Supercomputer/. The top500 list implies that a supercomputer is a system that can run Linpack really fast, while noting that the system must also be able to run other applications. http://www.top500.org/project/introduction

Given that NCSA has run many supercomputers over the years, and that I've personally run three while working there, I'd say that a good rule of thumb is that a supercomputer is a system designed to achieve high amounts of calculation throughput (as opposed to instant response) and that the system is at least 100x as powerful a high-end PC of that time frame. In fact, you could simplfy the rule down to- a system designed as a single unit to achieve high computing performance.

In order to accomplish all these things, supercomputers tend to have 2 things that "normal" network of PC's doesn't- a high speed, low latency network or interconnect, (and possibly several networks, each serving a different purpose) and a high speed, shared filesystem. Also, supercomputer tends to be designed and installed as a single unit, whereas a network of PC's happens over time.

Supercomputers tend to fall into one of 2 categories- a large collection of server class machines(cluster) or a small set of mainframe style systems(SMP). If you have the cash, you buy a large set of mainframe style systems, but who has the cash? Folks tend to purchase clusters as they tend to be less expensive, but you'd have determine if your application can work correctly on a large number of systems. Not all computing tasks can.

Tsubame, the system described above, is basically a cluster of inexpensive nodes with a high speed network. Applications on the cluster run on many of the individual nodes at the same time, and use the high speed network to pass messages to each other during the program, so that the application appears to be working on a single system. Tsubame is variant of a supercomputer cluster, where each inexpensive node is beefed up with co-processors and accelerators to increase the overall performance. Harder to program correctly, but potentially more powerful and still not as expensive as the large set of mainframes. Hope that helps.

The first time I read an article where I think Los Alamos was ordering a supercomputer with 8192 Pentium Pro processors in it, I was like WTF?

The system you're thinking of was called ASCI RED, and it was installed at Sandia, not Los Alamos:

http://www.sandia.gov/ASCI/Red/
http://www.top500.org/system/4428

The first time I read an article where I think Los Alamos was ordering a supercomputer with 8192 Pentium Pro processors in it, I was like WTF?

The system you're thinking of was called ASCI RED, and it was installed at Sandia, not Los Alamos:

http://www.sandia.gov/ASCI/Red/
http://www.top500.org/system/4428

The first time I read an article where I think Los Alamos was ordering a supercomputer with 8192 Pentium Pro processors in it, I was like WTF?

The system you're thinking of was called ASCI RED, and it was installed at Sandia, not Los Alamos:

http://www.sandia.gov/ASCI/Red/
http://www.top500.org/system/4428

"why use physical memory in modern systems"

Obviously, all computers use physical memory... duh :)
The question should be "why swap memory to disk in modern systems?"

The answer is that pretty much ALL performance-based systems (such as everything in the top 500 supercomputers, do not page. It is a performance versus convenience issue. Swapping is a huge convenience for most users, which allows large programs to load and run, yet not monopolize limited resources such as physical memory. If you do very little wrt running memory hog applications, then swapping will not be done, and will hardly affect you.

Nevertheless, equating virtual memory to page swapping on the front page of a geek journal was plainly wrong and should've been addressed.

How about #29 on the latest Top500.org list? Oh snap!

Check entry number 29. http://top500.org/system/9853 They're already there.

Oh Snap! How about #29 in the latest Top500 rankings?

your right, the Linpack benchmarks are indeed also measured in FLOPS, and that's what the TOP500 rankings are based on. however, when the Dawning is advertised as being capable of 180 TFLOPS or 160 TFLOPS--depending on who you ask--they're probably not referring to the Linpack benchmarks, which it only peaks at ~11.264 TFLOPS.

Problem is they've missed the boat. Linux already has compilers for multiple CPUs

Look at this chart..

http://www.top500.org/stats/list/32/os

Windows HPC 2008 is on 4 machines out of 500. (+1 is windows 2003 if you want to count that)
Linux is on 454 out of 500 super computers

Which Operating System do you think is going to have better tools to support Super Computing?

Problem is they've missed the boat. Windows already has all the tools needed for desktop computing

Look at this chart..

http://marketshare.hitslink.com/report.aspx?qprid=8

Windows has 90% of the market
Linux has less than 1%

Which Operating System do you think is going to have better tools for desktop computing?
__________________________

For all of you angry mods... i know it's a flawed analogy (and it's not even about cars...), but the parent was saying Microsoft should not even try since they're not at the top of the game. I think it's a pretty stupid way to think. Let them all try and develop new things. Someone will eventually come out with something that makes parralel computing easier and we'll all gain something from it even if it come from MS.