Events Calendar

Architectural Inference and the Pursuit of Efficiency

Benjamin Lee,
Stanford UniversityAbstract:
Energy efficiency is a defining challenge in modern computing. Limits in technology scaling pose challenges in power density while limits in software parallelism raise questions about future multi-core integration. Without process and parallelism to drive efficiency, we must rely on specialization and design coordination across the hardware/software interface. However, specialization is prohibitively expensive, incurring high non-recurring engineering costs that arise from an intractable number of degrees of freedom. I present the case for architectural inference to provide tractability for complex design questions in computer architecture. Inference enables comprehensive solutions to long-standing and previously intractable design priorities in heterogeneous specialization, application/architecture co-design, and architecture/circuit co-design. I also describe strategies for leveraging efficient components in cloud computing systems. In particular, I discuss experiences from deploying the Microsoft Bing search engine on mobile processors for energy efficiency. I also note the price of efficiency, which is exacted from application robustness and flexibility, and the implications for future system design. Biography:
Benjamin Lee is an NSF Computing Innovation Fellow in Electrical Engineering and a member of the VLSI Research Group at Stanford University. His research focuses on scalable technologies, power-efficient architectures, and high-performance applications. He is also interested in economics and policy for sustainable IT infrastructure. Dr. Lee has co-authored more than twenty papers in these areas, earning six nominations/awards such as the Harvard nomination for the ACM doctoral dissertation award, an IEEE Micro Top Pick, and a Communications of the ACM highlight. He has held visiting positions at Microsoft Research, Intel Labs, and Lawrence Livermore National Lab. Dr. Lee received his B.S. from the University of California at Berkeley and his S.M., Ph.D. from Harvard University.

Location: Hughes Aircraft Electrical Engineering Center (EEB) - -248

Audiences: Everyone Is Invited

Contact: Estela Lopez