By Kathryn Picardo
Computers are reaching their performance limit, Frances Allen of IBM told members of the College during her lecture on Wednesday, Oct. 19.
But Allen, a Turing Award-winning computer scientist, isn’t too worried.
“This is the best opportunity computer science has to improve user productivity, application performance and user integrity,” Allen said.
Allen won the Turing Award (roughly equivalent to a Nobel Prize in the computer science field) in 2006 after working with computers for more than 40 years. With IBM, she worked on the IBM Stretch, a machine that was meant to be 100 times faster than any existing product at the time, in 1956, and the Fortran Project, an initiative to enhance computers’ user productivity and performance, in 1954.
“Much of the algorithms and programming that we use today were built in the early years of the field,” Allen said.
Therefore, she continued, the solution to the present-day problem of peak performance capabilities lies in the innovations of a previous time.
“We need to enable bold thinkers and high-risk projects, like the early projects where it was not possible to repeat what has already been done,” Allen said. “We need to be willing to throw away some existing, familiar methods.”
The computer scientist’s apparent passion for her work inspired some students in attendance.
“It’s inspiring to see her achieve in the field she has dedicated her life to with such passion and commitment,” Ryan Manheimer, freshman physics and mathematics double major, said.