Dr. Valeria Bertacco
Arthur F. Thurnau Professor of Electrical Engineering and Computer Science
Associate Dean for Physical Sciences and Engineering
University of Michigan

Abstract
In the past 50 years, silicon dimensional scaling has been delivering the lion's share of advancement and economy of scale in computing. Unfortunately, this source of value is quickly drying out, highlighting the urgent need to provide novel approaches to accelerate innovation in computer design. In this talk, I outline three research domains that promise to provide effective alternatives to the once-plentiful benefits of silicon dimensional scaling. First, we now have the opportunity to move beyond silicon dimensional scaling and explore new silicon structures that boost system value in alternate ways. Second, we need to fully embrace specialized heterogeneous design, tailoring systems to upcoming powerful applications, thus leading to orders-of-magnitude performance and value benefits. Finally, and perhaps most importantly, we must break the cycle of ever-increasing design costs and diminishing innovation. By embracing novel low-cost and agile design technologies, through specialized programming languages and compilation flows, we can engage a broader community of engineers in accelerating innovation in computing system design.

Bio
Valeria Bertacco is Arthur F. Thurnau Professor of Electrical Engineering and Computer Science and Associate Dean for Physical Sciences and Engineering in the Rackham Graduate School at the University of Michigan. She is also an Adjunct Professor of Computer Engineering at the Addis Ababa Institute of Technology. Her research interests are in the area of computer design, with emphasis on specialized architecture solutions and design viability. She joined the University of Michigan in 2003, after working with the Advanced Technology Group of Synopsys, which she joined via the acquisition of Systems Science Inc.