Charles E. Leiserson is the Edwin Sibley Webster Professor in MIT EECS and a member of CSAIL. He leads the Supertech Research Group which works on software performance engineering, or “making code run fast or otherwise use few computational resources.” After coming to MIT 44 years ago, he designed a supercomputer which, in 1991, was the world’s most powerful computer. Now his group works collaboratively on tools to “make code go fast” for both experts and non-experts alike.
Professor Leiserson calls CSAIL, “My ideal place to work.”
Learn more about the Supertech Research Group’s work here.
A key goal of the Supertech research group is addressing the end of Moore’s Law, which, 1965, predicted that the number of transistors in an integrated circuit would double about every two years. This law, which guaranteed better computer performance over time, has been a driving force in the technology boom of the last 50 years. But Professor Leiserson says, “hardware efficiencies are no longer increasing at the rate that they used to.”
Now, without the assurance of increasingly effective hardware, Professor Leiserson’s group is digging into solutions that offer efficiency and performance gains and drive computing progress even if transistors can’t get any smaller. These solutions include parallel computing methods and creating easier ways to generate fast code. As the AI revolution gains steam, these researchers believe such work will offer critical answers to the demand for faster, cheaper, and better computing.
Learn more about Professor Leiserson’s thoughts on the End of Moore’s Law here.
