Nvidia names Stanford scientist its research chief

Nvidia on Wednesday named the chairman of Stanford University's computer science department as its new chief scientist, a particularly important position for the world's largest graphics chip supplier as it wages a technological war with Intel.

Bill Dally, who will be vice president of Nvidia Research, has been a professor of computer science at Stanford since 1997 and chairman of the computer science department since 2005. He will replace David Kirk, a renowned scientist in his own right, who will become an Nvidia fellow.

"Bill is legendary in the computer industry," said Jen-Hsun Huang, president and chief executive officer of Nvidia, in response to an e-mail query Wednesday. "He has made fundamental contributions, from parallel computing architectures to interconnects to low power designs to super fast I/Os" Huang said. "I expect him to contribute at all of those levels and more. And he will take forward David's work of building Nvidia research into one of the most regarded labs in the world."

Nvidia will need all the technological brain power it can muster in the coming years as Intel focuses its formidable resources on graphics technology like its upcoming Larrabee graphics chip. Nvidia has also faced increasingly pesky competition from Advanced Micro Devices' ATI graphics unit.

Like Nvidia, Intel is looking to designs that use increasingly sophisticated parallelism--large numbers of processors arrayed to do many tasks simultaneously. In fact, that's exactly what Larabee aspires to do. And Intel has been touting an experimental 80-core chip for a couple of years. Intel has also hooked up with DreamWorks to create animation using rendering farms that employ thousands of processors. Two Super Bowl commercials on Sunday will show off the fruits of this effort.

At Stanford, Dally and his team developed the technology that is found in many large parallel computers today, according to an Nvidia statement. At Caltech, he designed the MOSSIM Simulation Engine and the Torus Routing chip which pioneered "wormhole" routing and virtual-channel flow control. His group at MIT built the J-Machine and the M-Machine, experimental parallel computer systems.

Dally is a co-founder of Velio Communications and Stream Processors. He is a fellow of the American Academy of Arts & Sciences and also a fellow of the IEEE and the ACM and has received the IEEE Seymour Cray Award and the ACM Maurice Wilkes award. He has published over 200 papers, holds over 50 issued patents, and is an author of the textbooks Digital Systems Engineering and Principles and Practices of Interconnection Networks.

"Nvidia has always been on the hunt for top talent, and this is just another indication of that philosophy," said Jon Peddie, principal at Jon Peddie Research, which tracks the graphics chip industry. "The problem the U.S. and maybe the world face is that the computer industry is building millions of massively parallel processing chips in the CPUs and GPUs and almost no schools are teaching parallel processing programming," he said. "Either we get that changed ASAP or there's going to be a lot of cores sitting around twiddling their silicon thumbs." (CPU stands for central processing unit; GPU stands for graphics processing unit.)

(See Bill Dally's Stanford Web page here. "Current projects" are listed as "ELM: The Efficient Low-Power Microprocessor; On-chip Interconnection Networks; Sequoia: Programming the Memory Hierarchy; Scalable Network Fabrics.")

David Kirk has been with Nvidia since January 1997. He led Nvidia's graphics technology development for today's most popular consumer entertainment platforms. Kirk is the inventor of 50 patents and patent applications relating to graphics design and holds B.S. and M.S. degrees in mechanical engineering from the Massachusetts Institute of Technology and M.S. and Ph.D. degrees in computer science from the California Institute of Technology.