Silicon Investor (SI) -- The First Internet Community

STOCKTALK

We've detected that you're using an ad content blocking browser plug-in or feature. Ads provide a critical source of revenue to the continued operation of Silicon Investor. We ask that you disable ad blocking while on Silicon Investor in the best interests of our community. If you are not using an ad blocker but are still receiving this message, make sure your browser's tracking protection is set to the 'standard' level.

Technology Stocks : Intel Corporation (INTC) -- Ignore unavailable to you. Want to Upgrade?

To: Tony Viola who wrote (93417)	12/6/1999 11:08:00 AM
From: L. Adam Latham	Read Replies (1) \| Respond to of 186894

Tony: I don't think this got posted when the news came out a couple of weeks ago. The actual list is at top500.org. Adam sendmail.net Top 500 Supercomputers The new Top500 numbers are in, and your laptop has never looked so tragically slow. The Top500 report, a snapshot of the state of supercomputer installations worldwide, lists the sites of the world's 500 most powerful computer systems, ranked using the Linpack benchmark. The list has been updated twice a year since June 1993, and, in an field where biggerstrongerfaster is the metric of choice, has become the shorthand standard by which supercomputers are judged. No, Steve, your G4 didn't make the cut. Maybe next year. So who came out on top? Well, three US Department of Energy machines have taken spots one, two, and three to lead the list: ASCI Red (running on Intel processors) at Sandia National Labs in Albuquerque, ASCI Blue-Pacific (IBM) at Lawrence Livermore Labs in Livermore, and ASCI Blue Mountain (SGI) at Los Alamos. These are the only three systems to exceed 1 TF/s on the Linpack benchmark, and represent 7.4 percent of the total Flop/s on the list. A repeat winner, the upgraded Intel machine ASCI Red now boasts 2.38 TFlop/s. (In case you're curious, the Accelerated Strategic Computing Initiative, aka ASCI, is an initiative of defense programs at the DOE to "shift promptly from nuclear test-based methods to computation-based methods," and has nothing to do with 7-bit anything.) As of this rev, the threshold for top ten status has reached 0.56 Tflop/s. It's worth noting that the only Japanese system in this elite group is also the only (pseudo-)vector based system among them. Of the other nine, one is in Germany and the others are in the United States. To make it into this version's top 100, a machine had to do 101 GF/s or better. In fact, getting on the list at all isn't as easy as it used to be. The threshold for admission is now 33.09 GF/s, versus 24.7 GF/s last June and 17.1 GF/s a year ago. As a result, fully 222 of the machines on the June 1999 list have been pushed below the fold by more powerful systems - a replacement rate that's not only higher than the average churn of 160, but rates as the second-largest in the seven-year history of the Top500. Want more numbers? Sure you do. The total accumulated performance of the 500 systems on the list is 50.97 TFlop/s, up from 39.1 TFlop/s in June 1999 and 29.3 TF/s a year ago. Average growth rates per year are 1.8 percent for accumulated performance, 1.77 per year for the number one perch, and 2.0 per year for the number 500. This means the observed performance growth exceeds Moore's Law, which sets the bar at factor of 1.6 per annum. The new version of the Top500 list (the fourteenth so far) was published at the SC99 Conference in Portland. The next rev will go live at the Supercomputer 2000 Conference in Mannheim, held 8-14 June 2000. Start building your entry now.