To: Road Walker who wrote (125284 ) 1/17/2001 5:59:58 PM From: L. Adam Latham Read Replies (2) | Respond to of 186894 John and all: Don't know if this has been posted. Enjoy. Adamzdnet.com IBM's Building Fastest Linux Supercomputer By Peter Galli, eWEEK January 16, 2001 10:41 AM ET As the need for bigger, faster supercomputers increases at a rapid pace, IBM and the National Center for Supercomputing Applications at the University of Illinois at Champaign are working together to create the world's fastest Linux supercomputer. The NCSA said today it will install two Linux clusters, which include more than 600 IBM xSeries eServers running Linux and Myricom's Myrinet cluster interconnect network. IBM Global Services will install the first cluster next month, which will be based on IBM eServer x330 thin servers, each with two 1GHz Intel Pentium III processors running Red Hat Linux. The second cluster, which will run Turbolinux, will be installed this summer and will be one of the first to use Intel's next-generation 64-bit Itanium processor, IBM officials said. NCSA is the leading-edge site for the National Computational Science Alliance, a partnership to prototype an advanced computational infrastructure for the 21st century. It includes more than 50 academic, government and industry research partners from across the United States. Linux as the future Dan Reed, director of the NCSA and the National Computational Science Alliance, said Linux clusters provide users with a single, easy-to-use computing environment that applies to single-user desktop workstations, small research clusters and the largest "tera-scale" systems. "The explosion of the open-source community, the maturity of clustering software and the enthusiasm of the scientific community all tell us that Linux clusters are the future of high-performance computing," Reed said. Dave Turek, vice president of Deep Computing at IBM, said the Linux clusters would allow scientists to focus more on the results of their research initiatives, freeing them from the burden of building their own clusters and writing code to support their heavy computational demands. This is not Big Blue's first foray into supercomputing. Last year it put Linux on a 512-node cluster as part of the LosLobos supercomputer project at the University of New Mexico's Albuquerque High Performance Computing Center. LosLobos consists of 256 IBM Netfinity PC Servers. That 733MHz Intel IA-32 processor-based system is expected to provide a peak theoretical performance of about 375 gigaflops. The Albuquerque High Performance Computing Center also recently made the top 500 supercomputer list, with LosLobos the highest-rated Linux cluster on that list.