IBM and the US Department of Energy’s (DOE) National Energy Research Scientific Computing Center (NERSC) are to begin deploying the first systems on a computing Grid, which will allow researchers to tackle scientific challenges beyond the capability of existing computers.
The collaboration intends to make the largest unclassified supercomputer and largest data storage system within the DOE available via the Grid by December 2002.
The Grid will give scientists around the US access to far-flung supercomputers and data storage in the same way that an electrical Grid provides consumers with access to widely dispersed power-generating resources. This kind of seamless access to information is required for large-scale projects such as global climate, genomic and astrophysics research, which generate much more data than can be stored in a single location.
An emerging model of computing, Grids are built with clusters of servers joined together over the Internet, using protocols provided by the Globus open source community and other open technologies.
In the future, supercomputers, data storage and experimental facilities at Lawrence Berkeley, Argonne, Oak Ridge and Pacific Northwest national laboratories are expected to be connected to the DOE Grid.
NERSC itself is located at DOE’s Lawrence Berkeley National Laboratory, which has been developing distributed collaboration and distributed data handling technology for the past 10 years. This decade-long effort provided some of the precursor Grid tools and technologies.
In addition to a large IBM supercomputer system, Grid software will be integrated into NERSC’s HPSS (High Performance Storage System) archival data storage system, which has a capacity of 1.3 petabytes and is managed using IBM servers. By the end of the year, all three of NERSC’s IBM systems are expected to be on the Grid. To do this, IBM will develop its software to be compatible with Globus and other Grid software, and NERSC will then move the software into service.