The new system will employ the latest AMD Opteron central processing units as well as NVIDIA Tesla graphics processing units — energy-efficient processors that accelerate specific types of calculations in scientific application codes. The last phase of the upgrade is expected to be completed in late 2012. The system, which will be known as Titan, will be ready for users in early 2013.
With 299,008 cores and 600 terabytes of memory for solving critical problems, Titan will continue the legacy of its predecessor as one of the nation's most powerful tools for science.
The initial system upgrade will replace all of the processors with the latest AMD Opteron processors code named "Interlagos," taking the system from two six-core processors per node to one 16-core processor per node, and facilitate the addition of GPUs.
As one of the first deployments of the latest generation of AMD Opteron processors, the upgrade delivers one-third more cores in the same physical space compared with Jaguar while also doubling the memory and outfitting the system with Cray's powerful "Gemini" network to improve performance on scientific applications.
"But the game-changing new technology will be the inclusion of advanced GPUs," Mason said. "These Tesla GPUs are able to perform many more calculations for the same amount of power as conventional microprocessors and will work hand-in-hand with CPUs to deliver new levels of energy-efficient application acceleration."
Buddy Bland, project director for the upgrade, described the deployment timeline.
"In the first phase of the upgrade, to begin this year, we will replace all of the processor boards, memory and system interconnect network in Jaguar with the new Cray XK6 processor board. This upgrade will also include approximately 1,000 current NVIDIA Tesla 20-series GPUs," Bland said.
The GPUs will allow users to begin modifying their scientific applications to work well with the new technology. In the second phase, to be performed in the second half of 2012, between 7,000 and 18,000 of NVIDIA's next-generation Tesla GPUs will be added to the system, increasing the peak performance to between 10 and 20 petaflops, depending on architectural options executed.
If you liked this article, please give it a quick review on ycombinator or StumbleUpon. Thanks