Research Center for PlayStation 3 Processor Gains Continued Support at Georgia Tech

• By Dian Schaffhauser
• 07/11/08

The Georgia Tech College of Computing said that sponsors for one of its research centers have renewed their support. Sony, Toshiba, and IBM initially funded the STI Center of Competence with $320,000 in 2006 to do research into further uses for a new type of microprocessor that initially appeared in Sony's PlayStation 3 game console.

The Cell Broadband Engine (B.E.) processor, created through collaboration among the three companies, "supercharges" compute-intensive applications running in computer entertainment and handhelds, virtual-reality, wireless downloads, real-time video chat, interactive TV shows, and other "image-hungry" computing environments.

The STI Center has been responsible for creating and disseminating software optimized for Cell/B.E. systems, and for performing research on the design of Cell/B.E. systems, algorithms and applications. In conjunction with this renewal of the STI Center, the college announced a series of new research projects that are being undertaken at the center to develop applications and productivity tools based on the Cell/B.E. microprocessor.

"Today, we are carrying out the vision we always intended--to generate breakthrough innovations using Cell/B.E. technologies working hand-in-hand with researchers at Sony Group, Toshiba and IBM," said David A. Bader, professor and executive director of High-Performance Computing in the College of Computing. "We are very encouraged that our initial research results are showing the multi-faceted applicability of this technology."

One of the key research challenges that the collaborators will address through continued applied research is the use of Cell/B.E. technology to better monitor an aircraft's structural safety in commercial and military airplanes. Researchers will develop Cell/B.E. based data-processing software to monitor structural components in flight by measuring and recording an aircraft's vibrations through a distributed network of sensors. A commercial signal processing application for airplanes is a long term plan.

The other joint research projects include:

• A signal processing kernel needed for oil and gas exploration and seismic monitoring.
• Data compression, used for file compression or reducing the size of messages sent between computers.
• Financial services applications for consolidated debt optimization, as well as European and American options pricing.
• Encryption libraries for securing communications for privacy.
• High-speed multimedia codecs, such as MPEG2 and JPEG2000 encoders and decoders.
• Bioinformatics, such as DNA sequence alignment and comparison.
• Tools that involve a cross-platform profiler, performance estimation, and tuning system with IDE type features.
• A single-source automatic translator for generating PPU and SPU codes from a monolithic C/C++ application.