NSF plugs researchers into grid

The National Science Foundation on Aug. 9 approved a $53 million program to develop the world's largest computing grid dedicated to support open scientific research.

Four research centers will work primarily with IBM Corp., Qwest Communications International Inc. and Intel Corp. to build the Distributed Terascale Facility.

A high-speed "teragrid" network will connect the centers and enable scientists and researchers across the country — and eventually around the world — to share resources, scan remote databases, run applications on geographically dispersed computers and view complex computer simulations in real time from multiple locations.

The four research centers that will make up the DTF are the National Center for Supercomputing Applications (NCSA) at the University of Illinois at Urbana-Champaign; the San Diego Supercomputer Center (SDSC) at the University of California, San Diego; Argonne National Laboratory in Argonne, Ill.; and the California Institute of Technology in Pasadena.

When operational next year, the DTF network will be 16 times faster than the fastest high-speed research network available today. It is being described as the first step in transforming how scientific research is done. The teragrid will change the "way we live," said SDSC director Fran Berman. It will enable brain researchers to address fundamental ways of how the brain works, it will help researchers to simulate and view cancer drugs, and it will usher in a new era of human genome research.

The network will "transform" the researching world, said Dan Reed, director of NCSA and the National Computational Science Alliance. "The teragrid provides seamless access without people knowing where the infrastructure is and how it works," he said. "The DTF is the beginning of 21st century infrastructure for scientific computing."

Reed said he expects the grid to grow internationally to support international scientific collaboration.

The grid integrates the fastest supercomputers as well as high-resolution visualization environments and data storage facilities.

Traditional supercomputers are typically housed at a single location, but the grid infrastructure creates vast pools of computing resources by connecting widely-distributed supercomputers using the Internet or high-speed research networks. Organizations tap into these computing grids to access processing capacity, data storage and bandwidth just like a home appliance draws electricity from a power grid.

IBM Global Services will deploy clusters of IBM eServer Linux systems at the four DTF sites beginning in the third quarter of 2002. The servers will contain the next generation of Intel's Itanium microprocessor, code-named McKinley, said Robert Fogel, strategic marketing manager for Intel's high-performance computing.

IBM's supercomputing software will handle cluster and file management tasks, and Myricom Inc.'s Myrinet will enable interprocessor communication.

The system will have a storage capacity of more than 600 terabytes of data. The Linux clusters will be connected to each other via a 40 gigabits/sec Qwest network, creating a single computing system able to process 13.6 trillion calculations per second (13.6 teraflops).

The grid is gaining momentum in the research community, and it eventually will serve the needs of agencies' mission-critical systems, following a similar evolutionary path of Linux, said Michael Nelson, director of Internet technology and strategy at IBM.

"Three years ago, Linux was not ready for prime time," he said. "Security was not built in and basic functions were not provided, but in the last three years, those problems have been solved."