January 11, 2011

Scientific Community Behind Push to Exascale

Tiffany Trader

University of Tennessee Professor Jack Dongarra works at the forefront of supercomputing. In addition to being a professor of electrical engineering and computer science at the University of Tennessee, Dongarra also helps run the TOP500 list, a twice-yearly ranking of the 500 fastest supercomputers in the world.

A recent article at the University of Tennessee publication Tennessee Today portrays Dongarra as a champion of the exascale computing cause. Dongarra asserts that the next generation of computers will enable better solutions to national security, as well as make fundamental contributions to the economy, and advance competitiveness in the fields of engineering and manufacturing.

Today’s crop of computers has reached the petascale range. Jaguar, the famous supercomputer housed at the National Center for Computational Sciences in Oak Ridge, Tenn., operates at 1.7 Linpack petaflops (2.3 peak), giving it a second place finish on November’s TOP500 list.

The current level of computing power is sufficient for some scientific problems, but fails to provide clarity for others, leading many in the community to the conclusion that “science needs more power, more speed and more memory, or else scientific advancements will taper off.”

To this end, scientists have set their sights on the next level of computing, exascale, one-thousand times faster than petascale.

Says Dongarra:

“The drive to exascale is coming about from the science community. It’s not the technology that’s drawing us to exascale. The technology can take us there and that’s the good news, but it’s really the science that’s the driver, in some sense. These science applications have stepped up and said, ‘In order for us to do the kinds of problems that we can’t do today — in order for us to do them in the future — we need exascale computing.”

This challenge is so great that an international group of computer scientists have joined together to develop an exascale computing roadmap (PDF), with Dongarra at the helm. There are still cost and technology barriers that will need to be overcome, but the expert consensus predicts exascale computing will be a reality by 2020.

Full story at Tennessee Today

Share This