Slashdot Mirror

http://www.top500.org/list/2007/06/100

They not be .mil sites but I'm sure they do work for them.

qz

Because Storm was commissioned by Ballmer himself, after he saw the Top-500.
Quoth Ballmer:
"Fucking Top-500 supercomputers are fucking pussies. I'm going to fucking bury those machines, I have done it before, and I will do it again. I'm going to fucking kill those Superlamecomputers."
/ chair through window

Go, Windows! Take that, you Linux bitches!

According to this list you need a Rmax of a little over 4000Gflop to make it into the top 500 supercomputers these days...

#50 and #71 on the TOP500 list are composed of Apple Xserve G5s.

Now, Apple claims that the new quad-core Xserve with Xeon processors is five times faster than the Xserve G5.

Seems like building your supercomputer from Xeon Xserves would be a viable strategy for achieving a high rank on the TOP500 list!

...this is *hardly* a supercomputer. This is 152.57 times slower than entry number 500 on the Top 500 List. There isn't a nice neat definition of what a supercomputer is anymore, but "capable of running Beowulf" isn't it. Leaving aside the more custom machines that the company I work for (and a few others) build, there are plenty of Linux clusters that *do* qualify. The fastest one seems to be number 8 on the current Top 500 list (a Dell Infiniband cluster at NCSA).

With the relatively low cost and high availability of computing speed nowadays, the green500 list might become very important, as it is not only the environment-friendliness but also a lot of the running cost that is involved here.

I actually meant this, but that is actually 50% more powerful. Still, it's on the same order.

Typical short-sighted management only thinks in terms of what other short-sighted managers would do. They know that if they had a software company, they would use all their experience from their weekly readings of Joe Blow Manager Magazine or whatever to sell software the same way Microsoft does. They perceive all that as "business reality". Ironically, they trust adversarial behavior towards them such as attempts at vendor lock-in and one-sided licenses, because it is what they know. It sounds like normal "serious" business. So speak in "business reality" terms that they'll relate to. Turn perception upside down.

For example, there are two ways to make money from software, by selling it or by using it. Microsoft is in the business of selling software. Linux is used by a greater number of important companies because of the utility of better "quality control and customization" and greater "ROI". Now point your boss to the Top 500 List and then go to the statistics menu and sort by Operating System. Now show him how 69% of all the world's top supercomputers run Linux and how Windows only has 2 computers (0.4%) on the list. Apparently, Windows is completely inappropriate for high-end computing, and is only useful as a low-end platform for office productivity tools. From here you can point them to countless stastics of Apache's dominance of the web server market, how Google is a Linux shop, IBM is a Linux shop, the movie industry is basically standardized on Linux, etc.

i think my PS2 is supercomputer isnt it? Weren't the US government going to restrict exports on them as they were considered munitions or something daft like that. Same thing for old Mac G5 as i recall. Might be a stupid urban myth though

FYI Apple computers can be used to make a supercomputer. The MACH5 is number 50 in the TOP500.

Supercomputers are cheap, but many other products aren't getting cheaper at the same rate. A pack of toilet paper might be practically free if prices dropped as fast. However, who would stay in the toilet paper business. Computers will be cheap as long as money can be made. But there should come a day when anyone can make their own computer from raw materials, instructions posted online, and specialized power tools.

Having more powerful computers is tantamount to having more power. If personal computers become more and more powerful, catching up with yesterday's supercomputers as they exist today, people will be able to construct various items for themselves, including computers and machinery merely by selecting a few options and clicking OK. We'll wait with bated breath what happens to prices of common commodities.

If you search through the whole top500 list, you'll find these Ivy Leaguers with Blue Gene computers:

#93 Harvard
#382 Princeton

But, there are plenty of other US schools on the list with Blue Gene computers (and a many outside the U.S. as well):

#5 SUNY Stony Brook
#7 Renssellaer Polytechnic
#63 California-San Diego #374 Boston University
#376 Iowa State
#379 MIT
#383 Alabama-Birmingham

You're perfectly correct about the technology. It's your terminology that's wrong. When I said that mainframes are just a kind of microcomputer, I was not saying that mainframes are a kind of PC. "Microcomputer" refers to the fundamental technology that the PC is based on. "Mainframe" originally referred to computers built out of discrete components (the "mainframe" being that chassis that holds those components). At first that was the only kind of computer there was. Then the low end of the market was taken over by "minicomputers" which used integrated components. These were slower but cheaper. As time went by, minis got faster, and mainframes retreated into a high-end market.

Then around 1971, microprocessors appeared, people used them to make microcomputers, and history repeated itself. As microcomputers got faster and cheaper, they gradually took over the market. Completely. Now "mainframe" and "mini" describes market positioning, not fundamental technology.

Perhaps you're saying, "You're just playing with words. Mainframes are faster." But if so you're confusing "mainframe" with "high performance computer".

As it happens, I'm the documentation lead for an HPC system from Sun, the x4600. Up to 8 AMD CPUs. When Barcelona comes out, you'll be able to 32 processor cores in a single 4U system. Now, this thing has a fancy high-speed bus and other optimizations, but aside fro that, it's not that different from a PC. You can even run Windows on it. (Alas, most of our customers do, though Linux and Solaris are also popular.) You could even run an IBM emulator on it and then run all the software that was originally written for mainframes. Just imagine a Beowulf cluster of these!

There is the true 'big iron', like the top500. There are clusters and multi-core systems that use the exact same software technology. I've designed many such systems myself. Its those huge 1" tape drives and washing machine-sized Winchester drives that are no longer and these are the very things most associated with 'mainframes'.

It is the 'big iron' where Linux rules supreme with well over 70% share, almost 10x bigger than the second place systems. Go to http://www.top500.org/stats/list/29/os and see for yourself. Its no surprise most computing power is in Linux, but the lead is astounding.

Then there is Plan9, more of a 'networking system' than an 'operating system', but still considered Unix. This is one basis of 'super clusters' and its everywhere. If you tie thousands of machines saved from being 'PCs' into a cluster, that's a 'mainframe' if it looks like one or not. The 'mainframe' in whatever form it may take, has a bright future.

Think what you want, but I described the way things actually happened, while you're describing a gloss of revisionist history. Itanium was intended to kill off all other high end RISC development programs, and it very nearly succeeded despite being an "unsuccessful" and desperately late product.

SGI/MIPS canceled two high end CPUs, Beast and Capitan, specifically because of the threat of Itanium. I was there, I saw it happen.

http://news.com.com/Silicon+Graphics+scraps+MIPS+p lans/2100-1001_3-210024.html

Compaq killed Alpha before the HP merger, before Carly, with the intention of moving their high end business to IA-64:

http://en.wikipedia.org/wiki/DEC_Alpha

Obviously, Itanium redirected HP's focus away from PA-RISC since it was a HP/Intel project.

Itanium failed to completely derail SPARC, but it caused a great deal of controversy inside SUN about the future of the SPARC architecture and disrupted SPARC development for a year or two.

http://news.com.com/2100-1001_3-237583.html

IBM's Power architecture was perhaps the least affected by Itanium. IBM was pretty skeptical about Itanium and kept the Power program very much alive. As a result, they are the only RISC family which still has a significant presence on the Top500 supercomputing list.

http://www.top500.org/stats/list/29/procfam
http://www.top500.org/stats/list/13/procfam

I have no idea whether all these other CPU families would have been successful in the marketplace without Itanium. However, the fact is they were killed due to one of the most influential vaporware announcements in the history of the computer industry.

Think what you want, but I described the way things actually happened, while you're describing a gloss of revisionist history. Itanium was intended to kill off all other high end RISC development programs, and it very nearly succeeded despite being an "unsuccessful" and desperately late product.

SGI/MIPS canceled two high end CPUs, Beast and Capitan, specifically because of the threat of Itanium. I was there, I saw it happen.

http://news.com.com/Silicon+Graphics+scraps+MIPS+p lans/2100-1001_3-210024.html

Compaq killed Alpha before the HP merger, before Carly, with the intention of moving their high end business to IA-64:

http://en.wikipedia.org/wiki/DEC_Alpha

Obviously, Itanium redirected HP's focus away from PA-RISC since it was a HP/Intel project.

Itanium failed to completely derail SPARC, but it caused a great deal of controversy inside SUN about the future of the SPARC architecture and disrupted SPARC development for a year or two.

http://news.com.com/2100-1001_3-237583.html

IBM's Power architecture was perhaps the least affected by Itanium. IBM was pretty skeptical about Itanium and kept the Power program very much alive. As a result, they are the only RISC family which still has a significant presence on the Top500 supercomputing list.

http://www.top500.org/stats/list/29/procfam
http://www.top500.org/stats/list/13/procfam

I have no idea whether all these other CPU families would have been successful in the marketplace without Itanium. However, the fact is they were killed due to one of the most influential vaporware announcements in the history of the computer industry.

Here is the list of computers in use: http://www.top500.org/lists/2007/06/ and first SPARC is #178 http://www.top500.org/system/7522/

Here is the list of computers in use: http://www.top500.org/lists/2007/06/ and first SPARC is #178 http://www.top500.org/system/7522/

>>AFAIK, they never even touched the domain of supercomputing. But maybe I'm missing something?

Yes, you are not looking back far enough in the lists.
check out the lists from around 1998-2000
back then Sun's systems sometimes made up almost 25% of the Top 500 list
http://top500.org/stats/list/15/vendors/

The latest top 500 list is out today. http://www.top500.org/stats/list/29/osfam/ is the link to the OS statistics. It appears that Tamagochi farms aren't that popular amongst the high performance computing crowd.

You may find some insight at the top supercomputer list: http://top500.org/.
In particular, you can see that the supercomputers from 10-15 years ago are roughly equivalent in power to the average laptop on the market today.

between the professional corporate programmer and the console cowboy, gcc hacking, linux uber geeks

Reality shows us that the "console cowboy, gcc hacking, linux uber geeks" have produced, produce, and will continue to produce technologically SUPERIOR software when compared to anything produced by the "professional corporate programmer".

http://www.top500.org/stats/28/osfam/

"the most minimalist, zen-like, or spartan programming tools" are the very tools which have produced that which is more powerful than anything that CAN be produced with Visual Studio. See above link.

The "developer mindshare" among Linux geeks is comparatively low when weighed against, say the Windows platform and Visual Studio

Quantity is not synonymous with quality. It is those very point-and click-your-way-to-a-program "tools" such as VS which have made it possible for masses of folks (who would otherwise be in other fields) to piece together stuff for the windows platform.

I will leave the facts to be found at the above link as an exercise to the reader.

<fanboi mode on>

When you use the word "platform" I'm assuming we're talking about the OS itself, i.e. excluding the applications. I'm not very familiar with MS Windows but I think Linux is doing excellent qua performance and stability, witness that it can be used for embedded applications (thanks a lot to FSF's gcc and binutils, methinks), can be used for real-time (with modifications, but I thought they are going to be folded into the mainline kernel), is quite secure (selinux) and most of all is usable for serious computing: how many of the top-500 supercomputers run a kind of Linux (on at least some of the nodes)? At first glance I'd say about 70%. How many run *any* special edition of MS Windows, or other non-unix-like OS? At first glance I'd say 0%. Can you imagine what a boost this is for e.g. HIV and cancer research (paragraph 3 on the page)? Now imagine the real-life effects on society, if research centers were forced to use Microsoft software. To how many CPUs does that scale? Let's not even get started on "Windows for Warships" (for brits and maybe argentinos: listen to their sci-fi radio show -- but I digress).

<fanboi mode off/>

Of course that doesn't imply Linux is also a good desktop platform but I can't at the moment think of any OS feature that is specific for desktop use and that Linux can't provide. I may be a fanboi but yes, I'd say "Linux is better than MS Windows" (njaa njaa njaa etc.; penile length etc.).

Now how this translates to "has the best applications" is a completely different matter, for which technical excellence is much less important than inertia, portability of existing software, existing market share, and marketing (Microsoft marketing budget for Windows XP was $ 1 000 000 000 BTW; I'd say that compensates a lot).

Recently my University bought a supercomputer listed between the top500 computer systems in the world. During a class of Computational Physics, my Professor was commenting this issue and noted that the system was intended to serve about 80 research groups with tasks that demand parallel processing. The reality were much more modest though: only two groups were actually using the system (one of them was my Professor's research group), and of this two groups, only his group were taking advantage of parallel programming techniques to use the system the way it should be used.

He said that a new policy would be implanted, this was about 3 or 4 months ago, so I don't know how it is right now. But this reflects the lack of preparation to use such a system, they've wasted tons of cash with something they don't really know how to use properly.

Nice comment. It exactly express thw way we need to see OS: they are tools and we use wichever is the most usefull for our work. I would't use a screwdriver to repaint my new bedroom, as I wouldn't run a JBoss server on windows.

But I don't mind using windows to listen to DVDs or play games. Linux users tend to forget the whole OS is based on a SERVER OS, not a desktop. If they want to really be proud of their OS, they should go on http://www.top500.org/* and see the proportion of linux: THAT'S the linux achievement. THAT'S where it was supposed to be.

Now if they want to make it a proper OS for everyone as a desktop, they need to put the efforts so the average businesses see linux as a valid enough OS so they switch all their workstations to it. Then, with time, the same effect as with office/windows will happens: if you use an application at work/school, you'll want to use it at home because you feel confortable. For the games, the same effect will happen: the more user will have linux at home, the more game will be compatible.

Life is mostly a really slow snowball effect, but when the ball begin to get speed, it's hard to make them stop. We're still in the middle of the windows snowball effect, unless M$ begin acting stupid (wich I doubt), it will take lots of time to reverse the mouvement. Isn't it funny to see M$ FUD directly aimed at small business?

*From http://www.top500.org/: 2000: 5.60% share of linux 2004: 56,40% share of linux 2007: 75.20% share of linux

Nice comment. It exactly express thw way we need to see OS: they are tools and we use wichever is the most usefull for our work. I would't use a screwdriver to repaint my new bedroom, as I wouldn't run a JBoss server on windows.

But I don't mind using windows to listen to DVDs or play games. Linux users tend to forget the whole OS is based on a SERVER OS, not a desktop. If they want to really be proud of their OS, they should go on http://www.top500.org/* and see the proportion of linux: THAT'S the linux achievement. THAT'S where it was supposed to be.

Now if they want to make it a proper OS for everyone as a desktop, they need to put the efforts so the average businesses see linux as a valid enough OS so they switch all their workstations to it. Then, with time, the same effect as with office/windows will happens: if you use an application at work/school, you'll want to use it at home because you feel confortable. For the games, the same effect will happen: the more user will have linux at home, the more game will be compatible.

Life is mostly a really slow snowball effect, but when the ball begin to get speed, it's hard to make them stop. We're still in the middle of the windows snowball effect, unless M$ begin acting stupid (wich I doubt), it will take lots of time to reverse the mouvement. Isn't it funny to see M$ FUD directly aimed at small business?

*From http://www.top500.org/: 2000: 5.60% share of linux 2004: 56,40% share of linux 2007: 75.20% share of linux

You seem to have forgot one: Linux. If Linux weren't capable of using such "esoteric" hardware, why is it used on 75% of the world's 500 most powerful supercomputers, that most certainly use such hardware? At my place of work, we run several quite large Linux clusters, which use, among others, the Infiniband and Myrinet high-performance cluster interconnects. I can assure you that our vendors support those technologies on Linux, including 2.6.

Don't forget that 75% of the top 500 super computers (around 500K processors) run over that same non-existant OS. Really wonder where he got his "67% of the servers" stat. Especially since I don't know much company that would openly say "We currently hold X servers, Y run on windows and Z run on Linux/Unix".

http://www.top500.org/stats/28/osfam

Stop trolling, and though I shouldn't feed.. I will.

Where reliability isn't just important, it's critical. Where scalability isn't just important, it's critical. Where maintainability is valued over a hacker's OS because there aren't a bunch of free grad students to do all the damn work.

Hmm reliability, scalability, and critical workloads like perhaps with supercomputers? You'll note how Linux totally dominates this list with over 70% of all supercomputers. Where's Solaris.. oh that's right 1%. Also latest surveys have shown the majority of code commits to the Linux kernel as coming from major corporations like Novell, Red Hat, IBM. I will also say that you can't judge the code quality by the company behind it. I'd probably take most Hacker code over something written by some corporate drone who isn't passionate (as a hacker IS) any day. Grad Students want good code for thesis ;). Corporate employees want acceptable code to get through that 9 - 5.

Show me a Linux kernel that can handle multi-threaded apps running on 144 CPUs and using a terabyte of virtual memory.

What about this? 4096 Itanium2 Processors (64 Bit), 17TB of Ram. This system is multi-partitioned though, so it isn't all one kernel. However, they are using SUSE's Enterprise Server 9 bundled kernel which supports up to 512 Processors. So even there it's beaten your criteria for criticism.

Solaris has been fully 64-bit compliant for over a decade.

Linux has been 64bit for at least 7 years with Itanium and I'm assuming it has been 64 bit for over a decade with MIPS and Alpha architecture support. The majority of development was on i386 arch, however. I'm assuming this is now changing to x86_64 arch (like the majority of the world is running).

You're right, that's why nobody is using Linux for real systems.</sarcasm>

Except the shared bus the Xeons sit on is a seriously limiting factor, no-one in HPC is using Xeons because of it.
Thats funny, because according to the current Nov 2006 Top 500 Supercomputer list, there are about 220 Xeon systems (EM64T and IA32) on it.
I guess nobody told all those HPC professionals that nobody is using Xeons...

Except the shared bus the Xeons sit on is a seriously limiting factor, no-one in HPC is using Xeons because of it.

UT-Austin's current supercomputer isn't in the top ten right now, but "no-one" is a little harsh for number 12, don't you think?

I'm not saying that memory bandwidth isn't an important bottleneck (and I'd bet that's one reason they're going with AMD for TACC's next cluster), but depending on your application's behavior you can bring in enough work to keep two dual-core Xeons busy on each node, and I'm sure there are applications that won't starve two quad-core Xeons either.

Thank you for telling me that Earth Simulator went away 15 years ago!

From Anand, AMD, News.com and X-bit labs the numbers tell of an ever increasing marketshare for AMD. Note that this is for all servers, not just against Intel, meaning that Sun, IBM, and other servers are included. Then there's the 4 socket server category, where AMD has 48% of total marketshare in Q2 2006, and is the category I'd be more interested in as that is a true high-end server market. Then there's the Top 500 Nov 2006 List which lists 4 Opteron systems and a single Intel Itanium system in the top 10, along with 5 PowerPC systems. I should also note that only 31 Woodcrest systems are on the list vs 76 dual core Opterons.

Basically, AMD's server market share has been growing in leaps and bounds over the past 2 years, and has broken Intel's x86 monopoly in the space, especially once you exceed 2 processors. With Dell now finally offering AMD CPUs, I expect that number to grow.

The Sandia / Cray Red Storm super computer crammed inside of a Real Doll.

I build myself a top of the line machine about once a year, so I already have my "dream machine". But the above is a true dream machine. Especially if it washed your dishes, cooked your meals and ran linux.

D'oh! Fogot the the link :) http://www.top500.org/list/2006/11/100

Look at # 28, and notice which big company built the unit. That's right jacka**, it was built by a team of homegrown professionals from Huntsville, not IBM, HP, or Cray. The System Engineers built it, and it's beautiful.

I've worked in Supercomputing for almost 20 years and have yet to hear or see someone put an "s" at the end of anything when 1 is the unit. It's alway 1 Gigaflop, 1 Petabyte, 1 Megabit, etc, etc.

Look at http://www.top500.org/ and you'll see they don't put an "s" when referring to a single Teraflop.

We've run several PC clusters and IBM mainframes that didn't have a 1TF of capacity. You don't want know much power went into them. Yes, our modern blade-based clusters are more condensed, but they're still power hogs for dual and quad core systems.

Blue gene is considered to be a power efficient cluster and the fastest, but it still draws 7kw per rack of 1024 cpus. At 4.71 TF per rack, even Blue Gene pulls 1.5kw per teraflop.

Yes, it's a pair of video cards, and not a general purpose cpu, but your average user doesn't have ability to program and use a Blue Gene style solution either. They just might get some real use out of this with a game Physics Engine that taps into this computing power.

This is cool.

I don't think the GP meant the desktop. There are other sectors where Linux is much more prevalent. A good example is HPC, High-Performance Computing, where Linux powers 75% of the systems on the TOP500 list.

I just heard that talk; he gave it at EE380 at Stanford a few weeks ago.

First, this is a supercomputer guy talking. He's talking about number-crunching. His "13 dwarfs" are mostly number-crunching inner loops. Second, what he's really pushing is getting everybody in academia to do research his way - on FPGA-based rackmount emulators.

Basic truth about supercomputers - the commercial market is zilch. You have to go down to #60 on the list of the top 500 supercomputer before you find the first real commercial customer. It's BMW, and the system is a cluster of 1024 Intel x86 1U servers, running Red Hat Linux. Nothing exotic; just a big server farm set up for computation.

More CPUs will help in server farms, but there we're I/O bound to the outside world, not talking much to neighboring CPUs. If you have hundreds of CPUs on a chip, how do you get data in and out? But we know the answer to that - put 100Gb/s Ethernet controllers on the chip. No major software changes needed.

This brings up one of the other major architectural truths: shared memory multiprocessors are useful, and clusters are useful. Everything in between is a huge pain. Supercomputer guys fuss endlessly over elaborate interconnection schemes, but none of them are worth the trouble. The author of this paper thinks that all the programming headaches of supercomputers will have to be brought down to desktop level, but that's probably not going to happen. What problem would it solve?

What we do get from the latest rounds of shrinkage are better mobile devices. The big wins commercially are in phones, not desktops or laptops. Desktops have been mostly empty space inside for years now. In fact, that's true of most non-mobile consumer electronics. We're getting lower cost and smaller size, rather than more power.

Consider cars. For the first half of the 20th century, the big thing was making engines more powerful. By the 1960s, engine power was a solved problem, (the 1967 turbine-powered Indy car finally settled that issue) and cars really haven't become significantly more powerful since then. (Brakes and suspensions, though, are far better.)

It will be very interesting to see what happens with the Cell. That's the first non-shared memory multiprocessor to be produced in volume. If it turns out to be a dead end, like the Itanium, it may kill off interest in that sort of thing for years.

There are some interesting potential applications for massive parallelism for vision and robotics applications. I expect to see interesting work in that direction. The more successful vision algorithms do much computation, most of which is discarded. That's a proper application for many-CPU machines, though not the Cell, unless it gets more memory per CPU. Tomorrow's robots may have a thousand CPUs. Tomorrow's laptops, probably not.

IBM has had this for years. I was called the RS/6000 SP2. Great system. They have since replaced it with the Cluster1600 and with BlueGene/L Look at the top500 http://www.top500.org/ and you'll see they're up there.

-Aaron

While I can't give you exact numbers, when I went to a briefing this past summer, I was told that the Power6 chips would be about 50% faster than the existing Power5+ chips. Also, they would be multiple cores on a MCM (MultiChip Module) so that it appears to be an 8 core CPU.

If you want to see what kind of performance a Power5 and Power5+ processor can do, go look at http://www.top500.org/ 3 of the top 5 are Power5/Power5+ based and 4 of the top 5 are by IBM.

-Aaron

CISC has dominated RISC so much that 4 out of the top 5 computers in the top500.org list are RISC? (the one that isn't RISC is a Opteron Dual Core cluster) http://www.top500.org/list/2006/11/100 I had no idea that CISC was beating RISC so badly. As a side note, the 4 RISC systems in the top 5 were made by IBM. -Aaron

Sorry, didn't you notice? None of these run Windows. But I'm sure a CPU and OS emulator would run just fine.

If you look at the operating systems statistics , you can clearly see that the war is over and Linux has won :-)

In one camp, you have the guys that see Computer Science as a branch of Mathematics, and find it unfortunate that "Computer" appears in the name. For them Computational Science would be a much better name.

Unfortunately, the term "computational science" has now been taken by a third camp, which is completely different from the two that you mention. Computational science refers to the application of computers to solve scientific problems, typically by simulating physical phenomena. Computational scientists are what we traditionally think of scientists (e.g. chemists, physicists, climatologists, biologists, etc.) who do their work in computer simulation and use those big, expensive computers. For example, UCSD has a computational science program.

What about the IBM Blue Gene? It runs Linux, and currently Blue Gene/L has the largest Altix installation beat, as evidenced in the current Top 500 List

Yes, PowerPC is strong in the server room. Too bad for Freescale that 99% of them are IBM chips.

Presumably Freescale still has a decent share of the embedded market, but their position in general computing can be summed up in 8 characters: MPC8641D. Their amazing high-performance dual-core fast-FSB low-wattage super-G4 has been "just around the corner" since mid-2004.

If that chip and the 3GHz G5 had shipped on schedule, the results for http://www.google.com/search?q=boot-camp would be a lot different.

As a long time Folding At Home contributor, I found the following statement to be incorrect:

"Total power of their distributed computing network to 1-10 petaflops. At the upper end of that target, the network would be faster than any current supercomputer, at least in terms of FLOPS."

The fastest supercomputer in the world is IBM's Blue Gene/L, which clocks in at 280.6 teraflops http://www.top500.org/. The distributed network of Folding at Home is currently 171.2 teraflops http://folding.stanford.edu/stats.html/ as measured by sustained contributions from active members. The Folding At Home network is already the second most powerful, if ranked by the Top 500 list criteria. At any range from 1-10 petaflops, the Folding at Home network would be more powerful than the most powerful supercomputer in the world.

So which is yours?