Research
Energy-efficient computing
Enabling smaller, lighter, faster computers
It’s like starting over. Take all the algorithms you learned in your undergraduate class and throw them out the window…
Erik Demaine, professor
A theoretical computer scientist and his MIT colleagues are finding ways to reduce the energy used in computation—a change that could lead to laptops and mobile devices that are smaller and lighter, generate less heat, and perform complicated calculations with unprecedented speed. The researchers have proved mathematically that relatively simple hardware modifications could cut in half the energy consumed in running today’s standard software procedures. And they have shown that coordinated changes in software and hardware could increase the energy efficiency of computing by a million times. The researchers have already written new energy-efficient algorithms for everyday tasks such as searching and sorting that—when run on specially adapted computer hardware—should deliver substantial energy savings. And even greater savings will come with new energy-efficient procedures for processing big data, for example, during web searches.
Team
Research Team
Erik Demaine
Professor
Department of Electrical Engineering and Computer Science