ACM TechNews

The Multicore Challenge

Computing Community Consortium (08/26/08) Patterson, David

The jump to multicore processing is not based on a breakthrough in programming or architecture, but rather a retreat from the harder task of building power-efficient, high-clock-rate, single-core chips, writes former ACM president David Patterson, professor of computer science at the University of California, Berkeley. Numerous startup companies have tried to commercialize multiple-core hardware over the past few years only to be met with failure as programmers accustomed to improvements in sequential performance failed to adopt parallelism. If researchers manage to meet the parallel challenge, the future of IT will most likely be prosperous, but if not, failure could jeopardize both the IT industry and sections of the economy that depend on the rapid improvement of information technology. Such a failure could also provide an opportunity for the leaders in IT to move from the U.S. to wherever someone finds the solutions to writing efficient, parallel software. With such a crisis looming on the horizon, the Defense Advanced Research Projects Agency's choice to decrease funding for academic research in computer systems research since 2001 is troubling, Patterson writes. The lack of government support has driven industry to fund academic research efforts. Research efforts at Berkeley, Stanford University, and the University of Illinois are being funded by private industry, but it is unlikely private companies will spend more resources on such projects. Patterson argues that the U.S. government needs to return to its historic role of uniting great minds to solve important problems.

http://www.cccblog.org/2008/08/26/the-multicore-challenge/

---

© Copyright 2008 Information, Inc. This service may be reproduced for internal distribution.