The next generation of supercomputers, called exascale (a quintillion floating point operations per second), holds promise for solving some of the most demanding problems in numerical modelling. But as scientists worldwide, including Jack Dongarra, work to develop the hardware for these machines, the software to run them remains elusive.
Jack Dongarra recently received a $1 million grant, over three years, from the US Department of Energy to find a solution. Called the Parallel Runtime Scheduling and Execution Controller, or PaRSEC, his project aims to address the critical situation that is facing the supercomputing community due to the introduction of more complex supercomputer designs.
"You can't wait for the exascale computers to be delivered and then start thinking about the software and algorithms", stated Jack Dongarra. "The exascale computers are going to be dramatically different than the computers we have today. We have to have the techniques and software to effectively use these machines on the most challenging science problems in the near future."
Today's supercomputers have processor counts in the millions. Tomorrow's exascale computers will have roughly a billion processors. In addition, the general make-up of the machines will differ dramatically through the use of multiple central processing units and hybrid systems to overcome barriers limiting today's supercomputers. These barriers include large amounts of heat and power consumption, leaking voltage, and a limited bandwidth of data through the pins on a single chip.
With a goal of reaching exascale by 2020, the next generation of supercomputers promises to provide a broad range of industries, including energy, pharmaceutical, and transportation, the ability to more quickly engineer superior new products that could improve a nation's competitiveness. In addition, the advancements will translate into better consumer technology.
Jack Dongarra is also developing an algorithm to overcome a reliability problem associated with the increasing number of processors. Now, when one processor fails, the calculation may have to be repeated partially or in full. The project aims to develop software that can survive failures.
In addition to PaRSEC, Jack Dongarra is part of an international group working to evaluate potential rewards and obstacles in designing exascale supercomputers. Workshops called Big Data and Extreme-Scale Computing, hosted by the National Science Foundation, are held around the world annually. To learn more, you can visit www.exascale.org .
To learn more about PaRSEC, you can visit icl.cs.utk.edu/parsec .