Virginia Tech migrates supercomputer to Apple's new Xserve G5
Blacksburg, Va., Jan. 27, 2004 – Virginia Tech made supercomputing history in the fall of 2003 when it created the most powerful supercomputer at any university in the world in record time. Today, Virginia Tech announced plans to migrate its cluster of Power Mac G5 desktop computers to Apple's new Xserve G5 rack mounted 1U server. Xserve G5, the most powerful Xserve yet, delivers over 15 gigaflops of peak double-precision processing power per system and features the same revolutionary PowerPC G5, 64-bit processor used in Virginia Tech's cluster of 1,100 Power Mac G5s – the world's third fastest supercomputer.
"The Power Mac G5 based cluster validated our belief in the large-scale cluster architecture, the radically different communications technology, the 64-bit G5 processor and Mac OS X as a large-scale scientific computing platform," said Erv Blythe, vice president of information technology at Virginia Tech. "We know that the Xserve G5 cluster node running Panther Server will deliver even more impressive cost/performance numbers at our Terascale Computing facility due to its server optimized architecture, CPU density and ground-breaking performance and innovative management tools."
"Virginia Tech shocked the world of high performance computing by building the world's third fastest supercomputer based on the 64-bit performance of the PowerMac G5, it was a tremendous feat," said Philip Schiller, Apple's senior vice president of Worldwide Product Marketing. "Now they are advancing their breakthrough Terascale Computing facility to gain even more industry-leading price performance benefits by upgrading it to Apple's new 64-bit Xserve G5 cluster nodes."
Srinidhi Varadarajan, assistant professor of computer science in the College of Engineering at Virginia Tech, is the main architect of Virginia Tech's supercomputer called System X. A National Science Foundation CAREER Award recipient, Varadarajan developed a software package called Déjà vu that provides a fault tolerant software environment so that if any one component in the new supercomputer failed, the queuing system was alerted. Within milliseconds a free node takes over, averting the need to restart a calculation from scratch, a time frame that can potentially represent months.
The team that built System X, directed by Patricia Arvin, associate vice president of information systems and computing at Virginia Tech, and Glenda Scales, associate dean in the College of Engineering at Virginia Tech, will begin the transition immediately. It is expected to be completed by May.
Source: Eurekalert & othersLast reviewed: By John M. Grohol, Psy.D. on 21 Feb 2009
Published on PsychCentral.com. All rights reserved.