Keynote Speech I
Innovating in the Computational Intelligence Age
Senior Distinguished Research Scientist, NVIDIA
AMD Jerry Sanders Chair of Electrical and Computer Engineering
IBM-Illinois Center for Cognitive Computing Systems Research (C3SR)
Acting Department Head, Electrical and Computer Engineering
University of Illinois at Urbana-Champaign
Chicago, IL 61820
Wen-mei W. Hwu is a Senior Distinguished Research Scientist at NVIDIA and a Professor and the Sanders-AMD Endowed Chair Emeritus of ECE at the University of Illinois at Urbana-Champaign. He co-directed the IBM-Illinois Center for Cognitive Computing Systems Research Center (c3sr.com) 2016-2020 and the Intel-Microsoft Universal Parallel Computing Research Center (UPCRC) from 2006-2008. He served as a Principal Investigator of the NSF Blue Waters Supercomputer Project from 2013 to 2018. For his contributions, he received the ACM SigArch Maurice Wilkes Award, the ACM Grace Murray Hopper Award, the IEEE Computer Society Charles Babbage Award, the ISCA Influential Paper Award, the MICRO Test-of-Time Award, the IEEE Computer Society B. R. Rau Award, the CGO Test-of-Time Award and the Distinguished Alumni Award in CS of the University of California, Berkeley. He is a Fellow of IEEE and ACM.
Computational intelligence has been the driving force behind innovations in virtually all fields of the human society in the 21st century. We have been experiencing two very important developments in computing. On the one hand, a tremendous amount of resource has been invested into innovative applications such autonomous vehicles, recommender systems, and business/political analytics. On the other hand, the industry has been taking a technological path where traditional semiconductor scaling is coming to an end and application performance and power efficiency vary by more than two orders of magnitude depending on their parallelism, heterogeneity, and locality. With more than a decade of relentless advances in compute throughput and memory bandwidth, data movement has become the dominating factor for power, cost and performance of high-valued applications. It will be critical to match the data storage access and movement bandwidth to compute throughput and to locate the compute at where the data is. Much needs to be learned about of algorithms, languages, compilers and hardware architecture in this movement. What are the emerging killer applications that may become the new driver for future technology development? How hard is it to program existing systems to address the date movement issues today? How will we program future systems? How will the convergence of memory and storage devices present further opportunities and challenges in designing new systems? In this talk, I will present some lessons learned as we design the IBM-Illinois C3SR Erudite system at this exciting time.