The Defence Advanced Research Projects Agency (DARPA) is keen to give the US the edge in the supercomputer race by handing out money left, right, and centre in efforts to fund research and development of a superchip for the US military’s supercomputers.
DARPA awarded four different teams a research contract for the computer chip designed to increase GPU processing speeds by several thousand percent, choosing Nvidia as one of its partners due its extensive background developing graphics chips.
“This recognizes Nvidia’s substantial investments in the field of parallel processing and highlights GPU computing’s position as one of the most promising paths to exascale computing,” said Bill Dally, chief research scientist at Nvidia.
“We look forward to collaborating to develop programmable, scalable systems that operate in tight power budgets and deliver increases in performances that are many orders of magnitude above today’s systems,” he added.
The Nvidia team consists of Oak Ridge National Laboratory in Tennessee, supercomputer firm Cray in Seattle and a number of universities throughout the US.
While the large sum of money will be aimed at developing a supercomputer chip, the benefits of advances made may filter down to everyday users with new additions to Nvidia’s chip range.