Norman P. Jouppi, HP Computing Lab, USA


MPSoCs for Server Infrastructure


The “Big Data” explosion in business and science is creating the need for greatly increased server infrastructures. However, corporate and government IT budgets are not increasing significantly. This means that the total cost of server infrastructure (including energy costs) for a given amount of data processing must decrease dramatically. We believe MPSoCs will play a key role in enabling these big data applications with lower capital and energy costs. As MPSoCs begin to be deployed in these markets, many questions remain. For example, what is the best system architecture for these designs, including their mix of components and their specifications? How can networking costs be optimized? Do the benefits of MPSoC designs shrink or grow with further scaling? In this presentation we give initial results for these and other questions.


Norman P. Jouppi is an HP Senior Fellow and Director of the Intelligent Infrastructure Lab at HP Labs. He is known for his innovations in computer memory systems, including stream prefetch buffers, victim caching, multi-level exclusive caching and development of the CACTI tool for modeling cache timing, area, and power. Jouppi received his Ph.D. in electrical engineering from Stanford University in 1984, where he was one of the principal architects and designers of the MIPS microprocessor. He is a Fellow of the ACM and the IEEE, holds more than 60 U.S. patents, and has published over 100 technical papers, with several best paper awards.

* If you wish to modify any information or update your photo, please contact the web chairalain.fourmigue@polymtl.ca