New MacBook Pros are here! Get the lowest prices anywhere: Apple Price Guides updated Aug 1st (exclusive coupons)
 


Monday, October 25, 2004, 08:00 pm PT (11:00 pm ET)

Virginia Tech\'s System X achieves 12.25 teraflops

System X, the fastest supercomputer at any academic institution in the world, is now operating at 12.25 teraflops Virginia Tech said this week.

Virginia Tech's System X G5 supercomputer cluster originally ranked at 10.28 teraflops in the world's 500 fastest supercomputer. The new number represents an increase of almost two teraflops over the original System X, which consisted of 1100 dual 2GHz Power Mac G5s.

"Virginia Tech will learn of its new ranking when the list is unveiled in November of this year at SuperComputing 2004 in Pittsburgh," said Srinidhi Varadarajan, the lead designer of the system. "We expect to do well."

As part the cluster upgrade agreement with Virginia Tech, Apple agreed to provide 1150 special Xserve G5 servers to power the revised Supercluster. The custom built servers utilize dual 2.3GHz G5 processors and were developed specifically for the university. To date, Apple does not offer a 2.3GHz system.

According to Virginia Tech, the cost to rebuild System X was about $600,000, which included 50 additional nodes. By comparison, the original 1100 Power Mac G5 cluster rang up a $5.2 million dollar charge.

In addition to the companies the collaborated on the initial System X design– Apple, Mellanox Technologies, Emerson Network Power, and Cisco–Small Tree Communications, a Mac network solutions provider, has reportedly been instrumental in the operations of the rebuilt supercomputer. The company's communication software will be used to keep the clusters communications system current and up to date.

Virginia Tech initially sought the use of G5s to help establish that a radically different communications technology could be used to create a large-scale scientific computing platform. After proving its theory, the university approved a move to the Xserve G5 due to its server optimized architecture, computing power per unit density, and ground-breaking performance and innovative management tools.