Sunday, February 14, 2010

Supercomputers

PHOTO: The Columbia Supercomputer at NASA's Advanced Supercomputing Facility at Ames Research Center

Columbia Supercomputer at NASA is a cluster of 20 machines, each with 512 processors, each of which processes two data streams concurrently. Named in honor of the crew who died in the Columbia disaster, Columbia Supercomputer, built by Silicon Graphics for NASA, has as its main purpose to simulate the violent collision and merger of spiral galaxies that lead to the formation of elliptical galaxies. It is connected to the NASA Research and Engineering Network and was installed at the NASA Advanced Supercomputing facility in 2004.

Supercomputers were introduced in the 1960s and were designed primarily by Seymour Cray at Control Data Corporation (CDC). In the 1970s Cray established his own company, Cray Research and took over the supercomputer market with his new designs and held the top spot in supercomputing for five years (1985–1990). In the 1980s many competitors entered the market and many of these disappeared in the mid-1990’s supercomputer market crash. Currently supercomputers are custom-designed and produced by companies such as Cray, IBM and Hewlett-Packard.

Supercomputers are used for calculation-intensive tasks such as problems involving quantum physics, weather forecasting, climate research, molecular modeling such as computing the structures and properties of chemical compounds, biological macromolecules, polymers, and crystals, and physical simulations such as simulation of airplanes in wind tunnels, simulation of the detonation of nuclear weapons, and research into nuclear fusion.

As of November 2009, the fastest supercomputer in the world is the Cray XT5 Jaguar system at National Center for Computational Sciences with more than 19000 computers and 224,000 processing elements, based on standard AMD processors. The fastest heterogeneous machine is IBM Roadrunner, a cluster of 3240 computers, each with 40 processing cores and includes both AMD and Cell processors.

The SGI Altix platform was selected due to a positive experience with Kalpana, a single Altix 512-CPU system operated by NASA Ames which was integrated into the Columbia supercomputer system.

In November 2009, the AMD Opteron-based Cray XT5 Jaguar at the Oak Ridge National Laboratory was announced as the fastest operational supercomputer, with a sustained processing rate of 1.759 PFLOPS.

As against supercomputers, a quasi-supercomputer is like Google's search engine system with estimated total processing power of between 126 and 316 teraflops, as of April 2004. In June 2006 the New York Times estimated that the Googleplex and its server farms contain 450,000 servers. According to recent estimates, the processing power of Google's cluster might reach from 20 to 100 petaflops.

No comments: