Posted on 06/27/2007 8:09:14 AM PDT by ShadowAce
Call it "supercomputing smackdown." Today, at the International Supercomputing Conference 2007 (ISC07) in Dresden, Germany, the Top500 group announced its latest list of the top 500 most powerful supercomputers in the world.
To say it's a competitive list would be an understatement and few of the players can hold top spots long without constant improvements. With a few exceptions, this June's list has seen a reshuffling since the last list was published last November.
If there is such a thing as a winner in such a competition, however, that would be IBM (Quote). Just as last November – the top 500 list is recompiled every six months – Big Blue held the top spot, with its Blue Gene/L - eServer Blue Gene Solution. In fact, IBM has taken that top spot four times running with the system deployed by the U.S. Department of Energy (DOE).
"The Blue Gene/L System development by IBM and DOE’s National Nuclear Security Administration (NNSA) and installed at DOE’s Lawrence Livermore National Laboratory in Livermore, Calif., claimed the No. 1 spot," the Top500 group announced Wednesday.
That system achieved a blazing benchmark performance of 280.6 TFlops ("teraflops" or trillions of floating point operations per second) running what's called the Linpack benchmark. (For those who are not mathematically challenged, the Top500 group says the point of the Linpack benchmark is "to solve a dense system of linear equations.")
Interestingly, IBM's performance with the Power-based Blue Gene/L System was identical to its score in November 2006 and in November 2005. (IBM says the system's theoretical performance is 360 TFlops.)
Additionally, however, three other eServer Blue Gene systems made the top ten – last November, IBM only had two of the systems on the list. A fifth IBM system, a BladeCenter JS21 cluster system located at the Barcelona Supercomputing Center in Spain, and also Power-based, came in at number 9, down from the number 5 spot last November.
If that weren't enough, a sixth IBM system, an eServer pSeries p5 575, also made the top ten.
Despite such a strong showing, however, Big Blue doesn't completely dominate. Neither is it the only company whose supercomputers turned in teraflop performances.
Notably, Cray (Quote) supercomputers, powered by AMD processors, hold the number two and three spots on the new list. Both beat the 100 TFlop barrier with more than 101 TFlops apiece. Cray held the number two and ten spots last time.
Intel (Quote) chips are also represented in the top ten supercomputing sites with systems from Dell (Quote) in eighth place and SGI (Quote) at ten. Bumped from June 2007's top ten were Bull SA and NEC/Sun.
Begun in 1993, this is the 29th Top500 list to be published. A lot has changed since then. As a matter of fact much has changed in just the last six months.
"The performance needed to make it onto the list increased to 4.005 TFlops, compared to 2.737 TFlop/s six months ago, [and] the system ranked [at] 500 on the current list would have held position number 216 only six months ago .... the largest turnover rate between lists in the Top500 project’s 15-year history," the Top500 group's statement said.
Trend-wise, despite having only two in the top ten, Intel-based systems continued to gain ground as the predominant high performance computing (HPC) processor, with nearly 58 percent share – 289 out of 500. That's up from 52.5 percent last time.
AMD (Quote) came in second with 21 percent, a decline from 22.6 percent in November, while IBM Power-based systems represent 17 percent, or 85 systems, down from 18.6 percent six months ago. Dual core processors, clusters, and Gigabit Ethernet all dominate, although InfiniBand usage is growing for system interconnections.
The U.S. is home to 281 of the 500 systems on the latest list, and European sites have grown to 127 systems (a jump from 95), while Asian sites dropped to 72 (a decline from 79). Of those, Japan has 23 and China has 13 of the Top500 systems. The UK currently has the most in Europe -- 43 systems -- and Germany is second with 24.
One significant indicator of how dynamic the HPC/supercomputing market is: "The average age of a system in the Top500 list is only 1 year and 2 months," the group said. Also worthy of note, HP (Quote) systems did not rank in the top 50, while IBM systems comprise 46 percent of the top 50.
Can’t say I’m entirely surprised at that. I wonder what actual “work” is being done by the MS box?
The other day, in Costco, I passed a display of a 750 gigabyte drive, in a blister pack no less, for about $350!! That's 3/4 of a terabyte. Yikes....I'm old.
This is simply false, of course. HP has more supercomputers on the list than IBM, with IBM’s overall share dropping from 47 percent to 38 percent since the last report.
http://news.yahoo.com/s/pcworld/20070627/tc_pcworld/133515
106 is for show-off and probably testing. The other is apparently for corporate financial.
I just looked at the numbers for 106: 2048 processors = 1024 blades = 1024 copies of Windows * $469 each = $480,256 for just the OS in this lowly-placed cluster if a customer were buying, not counting any possible incentive discounts. Each of those blades costs about $4K = at least $4 million total for this system's hardware, not counting racks, enclosures, interconnects, etc.
Contrast with VT's achieving #3 for $5.2 million total in 2003, and its initial speed was still faster than this cluster is today.
I still remember the washer machines that you stuck the cake platters in. I kind of miss the sound of the punch card machine, too. But the paper tape was just weird.
I would imagine that Windows HPC is licensed somewhat differently than per-processor. I may be wrong, though. I haven’t looked into it.
I just looked up the per-copy price for WCCS, and know that Dell blade has two processors per. 2,048 processors in the cluster = 1,024 blades = 1,024 copies of Windows. This is probably plus or minus a bit depending on whether they count control nodes. There will also probably be some sort of volume licensing, but still that's a lot of money for just the OS when you get into thousands of servers.
Cheaper than 1024 copies of Windows...
I've been here a heck of a lot longer than you and you're still the biggest commie sympathizer I've ever seen. If you're not defending green party whackos you're defending Russians pirating American software and of course free technology for the Chinese government. You attack Christians about as good as most communists too, what did I leave out?
You forgot the yearly support costs for Linux. Sure the Russian and Chinese governments don’t pay that but most businesses do.
It could be programmed to reason, make logical inferences from available data, check their validity, learn about how it was right or wrong and then improve its algorithms used to make inferences. We already do self-modifying code.
I did say they would lack creativity. As I also said, it depends on your definition of intelligence.
And when it comes to computers, it's generally not a good idea to say "That's not possible." Someone will come along and prove you wrong as has happened many times before.
Just checked CDW, a single year of Red Hat Network (for security patches, only 30 days tech support) is $274 times 1,024 or $280,000 PER YEAR. That’s a million dollars every four years, or twice what Microsoft would have cost in that time. Not that I would recommwend Microsoft for this task but your whines about their cost is hilarious when compared to Red Hat.
As usual you don't know what you're talking about. Let's see if you can figure it out yourself this time.
Yea but you might pay support for Linux *might* many of the supercomputers running Linux out there. I mean some run SLES and some run ‘linux’. None of the first 122 are running redhat so I guess its very possible that they are not paying for support..
I should have known better than trust your figures for Microsoft’s cost as well, the price you quoted of ~$450 per is three times the lowest price on CDW of $150. The Red Hat price I gave of $274 is the lowest I saw, and that’s still PER YEAR pricing.
Then again now that I think about it if youre going to buy support for Linux why would you not buy it from MS as well so with MS you get the support and license cost..
“I was a member of the team that installed #17 at ARL this year.”
Thank God for conservative Linux/Unix users. With all of the Stallmanites out there, I’d go mad without you guys.
“I am excited by the rapid increase in capacity and performance and am looking forward to the day when AI systems become smarter than us despite their linear limitations. By then, I hope, we will have the means of linking our “holistic” brains with theirs and get the best of both worlds ...”
Check out Quantum Computing. In 20-30 years we will all be connected, one global conciesness. Weather will be able to be forecasted out with 99% accuracy 50 years in advance. I also wager we will make break-throughs to insert AI into protons, accelerate them faster then light (this part has been accomplished) and map our entire universe by 2100. Forget Star Trek for space travel, eventually it will be human consciessness inserted into the proton. Of course, that means we have evolved at that point and I wager we will then find many, many forms of intelligent beings whom have already reached such a phase.
What is human intelligence? Electonic impulse to command our motor functions and to trigger other primal needs to sustain our body. Other then that you have the subconsciense which is basically and advanced hard drive to store and access memories and attempt to solve problems which the conscience cannot comprehend. Einstein like in many areas probably has this one right as well.
Disclaimer: Opinions posted on Free Republic are those of the individual posters and do not necessarily represent the opinion of Free Republic or its management. All materials posted herein are protected by copyright law and the exemption for fair use of copyrighted works.