HP, Nvidia Team Up in HPC Project

Hewlett-Packard and Nvidia are providing key technologies to a project aimed at designing an experimental high-performance computing system.

The two technology vendors are part of a project being overseen by the Georgia Institute of Technology, which announced Oct. 21 that it had received a five-year, $12 million Track 2 award from the National Science Foundation’s Office of Cyberinfrastructure.

The goal of the program is to create two heterogeneous HPC systems for research work in such areas as computational biology, combustion, materials science and massive visual analytics.

In the initial system, which is scheduled to be deployed in early 2010, HP and Nvidia will provide the processing power and computing systems. The project will pair the CPU capabilities of Intel with GPUs (graphics processing units) from Nvidia.

HP will supply hundreds of Intel-based systems, while Nvidia will bring its next-generation CUDA architecture, which is code-named “Fermi.” CUDA is the computing engine for Nvidia’s GPUs. CUDA also makes it easier for researchers to run GPUs and CPUs in a co-processing way. Fermi is designed specifically for HPC environments.

Program designers say the program will be the first of the Track 2 awards—which are used for designing leading-edge computing systems that run near a petascale level—to leverage GPUs.

Click here to see how GPUs can help in HPC environments.

Nvidia and rival Advanced Micro Devices have been aggressively pushing GPUs into more mainstream computing environments, noting their ability to scale and to drive down energy consumption. Bill Dally, chief scientist at Nvidia, said the combination of GPUs and CPUs is particularly powerful.

“Computational science is a key area driving the worldwide application of GPUs for high-performance computing,” Dally said in a statement. “GPUs working in concert with CPUs is the architecture of choice for future demanding applications.”

Nvidia officials in September announced they were partnering with Microsoft to promote the use of their Tesla GPUs in HPC environments that use Microsoft’s Windows HPC Server 2008 operating system.

Ed Turkel, manager of business development for HP’s Scalable Computing and Infrastructure business unit, said the combination of Intel CPUs and Nvidia GPUs will lead to increased performance and faster application development.

“Research institutions are looking for energy-efficient, high-performance computing architectures that can speed time to solutions,” Turkel said.

Also involved in the project are researchers from Georgia Tech’s College of Computing, Oak Ridge National Laboratory, the University of Tennessee, and the National Institute for Computational Sciences.

“Our goal is to develop and deploy a novel, next-generation system for the computational science community that demonstrates unprecedented performance on computational science and data-intensive applications, while also addressing the new challenges of energy efficiency,” Jeffrey Vetter, joint professor of computational science and engineering at Georgia Tech and Oak Ridge, said in a statement.