INDUSTRY

SANTA CLARA, Calif., - More than one-third of the world's most powerful computers with Intel processors inside of them are in commercial use, signaling that high performance computing (HPC) is evolving from its academic origins and into mainstream business. The list of the world's 500 most powerful computers, issued today and in conjunction with the International Supercomputer Conference in Heidelberg, Germany, also reveals that affordable systems with as few as 92 Intel processors are appearing on the list for the first time. These systems are delivering better performance vs. many proprietary systems at lower costs. "HPC is going mainstream," said John Davies vice president, Intel Sales and Marketing Group. "Businesses like Shell and Lockheed Martin are joining leading academic institutions and research labs in using powerful, standards-based Intel processors and software to tackle the world's most complex computing challenges. With these powerful modeling and research systems, they can gain a tremendous business advantage." Intel-based systems on the elite TOP500 list totaled 119, more than double the 56 systems on the list only six months ago, and up dramatically from just two on the list three years ago. Nineteen Intel® Itanium® architecture-based systems appeared on the list, up from two last time. The Intel® Itanium® 2 processor family also broke into the top 10 for the first time. An HP system in use at the Pacific Northwest National Laboratory in Richland, Wash. secured the eighth position on the list. Notably, the Intel® Xeon(tm) processor-based system cluster at Lawrence Livermore National Laboratory achieved No. 3 on the list, the highest position ever for any cluster. It uses a Quadrics interconnect and was manufactured by Linux Networx. Cluster systems such as this are also being discussed this week at the ClusterWorld show in San Jose, Calif. Intel: HPC and Grid Computing Today, Intel processors, platform architectures, interconnects, networking technology, software tools and services power some of the fastest computers in the world at price points that have expanded HPC beyond the confines of supercomputer centers and into the broad community of users in mainstream industries and influential scientific, research and academic organizations. Grid computing takes the HPC philosophy and extends it by linking desktops, clusters and large symmetric multiprocessing systems across multiple geographic locations - creating a single, virtual computing resource. This enables the integrated, collaborative use of high-end computing systems, networks, data archives and scientific instruments that are operated and accessed by multiple organizations.