acm-header
Sign In

Communications of the ACM

ACM TechNews

Nationwide Research Network Will Expand Supercomputer Capabilities


View as: Print Mobile App Share:
Indiana University President Michael A. McRobbie

"This project is so vitally important to the future of scientific research at all institutions of higher learning," says Indiana University President Michael A. McRobbie.

Credit: Chris Meyer / Indiana University

Indiana University (IU) researchers will lead a four-year, $15 million National Science Foundation project to develop software to connect future supercomputers and enable new approaches for scientific research on massively large problems. The grant will be used for the construction of an experimental supercomputing network called FutureGrid, which will consist of nearly 1,400 advanced computer processing units spread across six universities.

"The ultimate goal of this project is to create the next generation of investigative tools for scientific researchers whose computational needs often exceed the capabilities of a single institution or network," says Indiana University president Michael A. McRobbie.

The project aims to make it easier for scientists to conduct research that requires enormous data processing capabilities, such as the complex modeling of climate systems or analyzing and comparing DNA segments and complex organic molecules. IU's Brad Wheeler says a variety of supercomputers will be acquired for the network. "Each type of supercomputer has a unique architecture and capabilities that make it ideal for certain types of uses," Wheeler says. "One of our goals in this project is to learn by conducting formal experiments for the best ways to put all these computers together for researchers."

Wheeler says the FutureGrid will experiment with connecting supercomputers in a variety of ways to determine which connections and software combinations work the best. The goal is to create a system researchers can use for supercomputing projects without having to worry about hardware design and capabilities.

From Indiana University
View Full Article

 

Abstracts Copyright © 2009 Information Inc., Bethesda, Maryland, USA


 

No entries found

Sign In for Full Access
» Forgot Password? » Create an ACM Web Account