University of Cambridge > > Computer Laboratory Computer Architecture Group Meeting > The CAVA Computer: Exceptional Parallelism and Energy Efficiency

The CAVA Computer: Exceptional Parallelism and Energy Efficiency

Add to your list(s) Download to your calendar using vCal

If you have a question about this talk, please contact Prof Simon Moore.

Note unusual time and venue

Abstract: Designing a new computer system is a very expensive proposition. But 80% of it is exactly the same as every other computer-you need caches, multiprocessing, coherency protocols, memory systems, etc. Getting all of that right requires skill and experience, but is taken for granted and does not command much of a premium. Getting it wrong, on the other hand, is a commercial disaster. In this talk I propose a research initiative to standardize and open-source a design for the 80% that is the same in every design, so that everyone can concentrate on adding value to their own remaining 20%. The CAVA computer is a “cluster in a rack” energy-efficient parallel computing architecture targeting 10nm CMOS technology. The first part of the talk describes a 1024-node system where each node consists of 96-core, 3-issue out-of-order processor chips running at 1GHz with four DDR4 memory channels. Power estimates of different components are discussed, as well as cost projections. The second part of the talk discusses architectural tradeoffs that were made, how this architecture might play in the HPC exa-scale arena, and broader market implications. The talk concludes with how I envision the simulation infrastructure is organized, what Oracle Labs brings to the table, and a list of research topics that I and others at Oracle Labs are actively researching and would be interested in working with students at Universities. I hope to organize an in-depth research effort into designing this computer and, if sufficient progress can be made, perhaps building a prototype.

Bio: Peter Hsu was born in Hong Kong and came to the United States at age 15. He received a B.S. degree from the University of Minnesota at Minneapolis in 1979, and the M.S. and Ph.D. degrees from the University of Illinois at Urbana-Champaign in 1983 and 1985, respectively, all in Computer Science. His first job was at IBM Research in Yorktown Heights from 1985-1987, working on superscalar code generation with the 801 compiler team. He then joined his ex-professor at Cydrome, which developed an innovative VLIW computer. In 1988 he moved to Sun Microsystems and tried to build a water-cooled gallium arsenide SPARC processor, but the technology was not sufficiently mature and the effort failed. He joined Silicon Graphics in 1990 and designed the MIPS R8000 TFP microprocessor, which shipped in the SGI Power Challenge systems in 1995. He became a Director of Engineering until 1997, then left to co-found his own startup, ArtX, best known for designing the Nintendo GameCube. ArtX was acquired by ATI Technologies in 2000, which has since been acquired by AMD . Peter left ArtX in 1999 and worked briefly at Toshiba America, then became a visiting Industrial Researcher at the University of Wisconsin at Madison in 2001. He then consulted part time at various startups, and attended the Art Academy University and the California College of the Arts in San Francisco where he learned to paint oil portraits, and a Paul Mitchell school where he learned to cut and color hair. In the late 2000’s he consulted for Sun Labs, which lead to discussions about the RAPID project post acquisition. Peter joined Oracle Labs as an Architect in 2011.

This talk is part of the Computer Laboratory Computer Architecture Group Meeting series.

Tell a friend about this talk:

This talk is included in these lists:

Note that ex-directory lists are not shown.


© 2006-2022, University of Cambridge. Contact Us | Help and Documentation | Privacy and Publicity