Stanford Gets $6M to Research Parallel Processing; Aims at Virtual Worlds

Last month saw Berkley nab $10 million each from Microsoft and Intel to research coding for parallel processors. Now Stanford is set to announce its Pervasive Parallelism Lab with $6 million in funding over three years from Sun Microsystems, AMD, Nvidia, IBM, HP, and Intel. Most new computers already feature multi-core processors, but not all software (or, it often seems, very much of it) is ready to make use of them. Stanford's general goal is to improve programming techniques, but the head of the lab, Kunle Olukotun, says there are three specific areas of focus: virtual worlds, robotics and massive data analysis jobs.

"You might have a very heterogeneous mix of hardware and software coming together for the same application," Olukotun told the Channel Register. "A video game can manage virtual objects through dynamic threading but also use streaming techniques for physical world modeling. Then, there might be a 3-D graphics component that requires its own form of parallelism."

Nvidia has shown some extra interest in virtual worlds already, and IBM and Intel have both partnered with HiPiHi to help squeeze extra power out of users' hardware. That, of course, is in addition to plenty of other work the two are actively engaged in with other partners. And, not to short any of the partners, Sun, AMD, and HP have also had fairly substantial activity in the virtual worlds space, including Sun's own Wonderland and Darkstar programs for creating and hosting virtual worlds.

As the industry looks to the future, there's always a focus on when the average consumer will have a robust enough hardware setup to run the immersive worlds of today, not to mention the richer applications of tomorrow. Hopefully Stanford's program will help create programs that are ready for that hardware.

"We believe in driving applications,"  Hanrahan said in a statement. "Among the most interesting are immersive, richly graphical, virtual worlds, both because of the unique experiences for users as well as the challenges in building such demanding parallel applications."