“We have reached the limit of what is possible with one or more traditional, serial central processing units, or CPUs,” says Bill Dally, chief scientist and senior vice president of research at NVIDIA, citing the failure of power scaling (energy consumed by each unit of computing would decrease as the number of transistors increased).

“It is past time for the computing industry–and everyone who relies on it for continued improvements in productivity, economic growth and social progress–to take the leap into [energy-efficient] parallel processing.”

The problem: “Converting the enormous volume of existing serial programs to run in parallel is a formidable task, and one that is made even more difficult by the scarcity of programmers trained in parallel programming.”