Industry


Ads by TechWords

See your link here


IT Blogwatch's picture
IT Blogwatch

A Daily Digest of IT Blogs from Richi Jennings

IBM's 1.6 PFLOPS supercomp. (and LAMP consultant joke)

You're welcome to IT Blogwatch, in which we learn what IBM's doing with Sony's surplus PS3 processors. Not to mention a real-life unqualified consultant joke...

Steven Schwankert schwaggers in with this:

The U.S. Department of Energy (DOE) has chosen IBM to build a new supercomputer for its Los Alamos National Laboratory in New Mexico, IBM announced yesterday. Named Roadrunner, the machine comes with a price tag of $35 million.
...
DOE has used supercomputers to keep an eye on the health and readiness of the country's nuclear weapons, in lieu of actual underground nuclear tests ... will run Red Hat Inc.'s Version 4.3 Linux operating system and will be built entirely with commercially available hardware running two different types of processors ... Advanced Micro Devices Inc.'s Opteron processor, alongside ... Cell processors.

Thomas Ricker clarifies the Cell angle, but kinda loses it at the end:

One reason there's so much fuss (and delay) over the upcoming PlayStation 3 platform is the fact that it sports those new Cell processors jointly developed with Sony and Toshiba. Now those Cell procs are about to find themselves pumping away at the heart of a new $35 million supercomputer ... IBM is hoping to reclaim the title of the world's fastest once completed in 2007. Dubbed Roadrunner, ACME IBM plans to jump from 280-teraflops to a full petaflop performance by combining AMD Opteron blade servers and Cell-based accelerator systems. A performance threshold achieved earlier this year by NEC's MDGrape-3, the supercomputer behind new pharmaceutical drugs and the curious taste and powerful punch of Mad Dog's 20/20 Red Grape Malt beverage. Mmm, grape.

Hubert and Eliane Nguyen blogged this from their kitchen:

IBM is at it again, trying to build the world's most powerful supercomputer at a US government laboratory that could potentially be up to 4 times as fast as IBM's very own BlueGene/L ... It is hoped that the Roadrunner will hit petaflop speeds, maxing out at 1.6 thousand trillion calculations per second. In comparison, the BlueGene/L is only capable of a trillion calculations per second. The Roadrunner has a projected completion date of 2008 and will take up approximately 12,000 square feet of floor space.

Greg Stierle adds this:

This news is significant for a few reasons. One it speaks to the importance of the work IBM has done with chips for non-traditional uses. By being willing to broaden their horizons, or lose focus, depending on what business school trend is hot, IBM has been able to enhance a core competency.

A second reason this is significant to me is that it demonstrates how IBM can take advantage of advances in one area of its business and monetize it in another. I know of no other company that does as good of a job with monetizing research and patents, and that experience worked to IBM’s advantage here as well.

The third reason I find these sorts of announcements worthwhile, is that it cements IBM in the public mind as being uniquely competent in certain areas of computing.

Slashdotters discuss:

Cybert4: they gloss over whether they mean floating point operations or "calculations" per second. The article seems to equate a flop with "calculations per second". The flop, of course, came from floating point operation. Even then it's vague--is it single, double or double-extended? [Dude, it's a FLOPS not a FLOP]

Junta: Definitely flops. Though not necessarily 64-bit precision flops, as are required for top500 scores... The cell isn't impressive double-precision wise.

scoobrs: no Cell processor ever reaches the impossible speed of 360 GFLOPS on any real world scientific application because of the real world problems of a slow interface to memory, storage, network, etc. ... When numbers are being used in a press release, all vendors in the industry benefit greatly from using whichever number is the largest and most impressive to the reader, even if it is completely impractical to a supercomputer user.

syntap: And looking back 20 years from now we'll laugh at such a large room full of computer equipment, the equivalent of which will be powering our mobile communications devices in a 150mm x 150mm package.

Buffer overflow:

Around the Net

Around Computerworld

And finally... How Many Consultants Does It Take To Turn On A LAMP?

Richi Jennings is an independent technology and marketing consultant, specializing in email, blogging, Linux, and computer security. A 20 year, cross-functional IT veteran, he is also an analyst at Ferris Research. Contact Richi at blogwatch@richi.co.uk.

What People Are Saying

I wonder for what they need

I wonder for what they need so powerful computers? Are they still calculating nuclear bombs range?

I get the feeling that that

I get the feeling that that none of those colourful opinions come from people who actually program scientific software on a daily basis. I do, I also write scientific cluster software for CPU intensive algorithms. My point is, if they are going to make these little jokes and comments, they should at least KNOW fully what they're talking about (they clearly do not), it gets embarressing otherwise.

Theres a great quote from

Theres a great quote from Sandia National Labs: There are lies, damn lies, and benchmarks. Another general heuristics is: all models are inaccurate, some models are useful.

A Petaflop is revving the engine, its not the same as running a few laps around the track (process based performance benchmark), or actually running the race.

Surplus Cell processors

Surplus Cell processors eh?... good one.