Nvidia to power world's fastest supercomputer

Cray XK6 'Titan' big iron with 18,000 Tesla GPUs
Author:
Publish date:
4_nvidiacray184.jpg

Graphics specialist Nvidia announced that the Oak Ridge National Laboratory will deploy a new Nvidia Tesla GPU-based supercomputer called Titan which will be over twice as fast as the current world's fastest supercomputer.

Titan, a Cray XK6 supercomputer, will also be over three times more energy efficient than the world's fastest today, the Japanese K supercomputer.

ORNL operates an open science computing facility for the US department of energy and Titan will be used for research in a broad range of fields, including material science, energy technology, medical research, geoscience and more.

"Oak Ridge's decision to base Titan on Tesla GPUs underscores the growing belief that GPU-based heterogeneous computing is the best approach to reach exascale computing levels within the next decade," said Nvidia Tesla boss Steve Scott.

"The Tesla GPUs will provide over 85 percent of the peak performance of Titan. You simply can't get this level of performance in a power- and cost-efficient way with CPUs alone," he added.

The first phase of the Titan deployment is already underway with Oak Ridge upgrading its existing Jaguar supercomputer with 960 Tesla M2090 GPUs. These GPUs will serve as companion processors to multi-core CPUs in the upgraded Cray XK6 supercomputer.

In the second phase, expected to begin in 2012, Oak Ridge plans to deploy up to 18,000 Tesla GPUs based on the next-generation architecture code-named "Kepler." Titan is ultimately expected to deliver computing performance in excess of 1,000 petaflops.

Related