Thursday, October 6, 2022 - 4:30pm to 6:00pm
Location:Group Viewing and Virtual Presentation -ET Rashid Auditorium, Gates Hillman 4401 and Remote
Speaker:DAVID A. PATTERSON, Pardee Professor of Computer Science, Emeritus University of California, Berkeley https://www2.eecs.berkeley.edu/Faculty/Homepages/patterson.html
A Decade of Machine Learning Accelerators: Lessons Learned and Carbon Footprint
The success of deep neural networks (DNNs) from Machine Learning (ML) has inspired domain specific architectures (DSAs) for them. ML has two phases: training, which constructs accurate models, and inference, which serves those models. Google’s first generation DSA offered 50x improvement over conventional architectures for inference in 2015. Google next built the first production DSA supercomputer for the much harder problem of training. Subsequent generations greatly improved performance of both phases. We start with ten lessons learned, such as DNNs grow rapidly; workloads quickly evolve with DNN advances; the bottleneck is memory, not floating-point units; and semiconductor technology advances unequally.
The rapid growth of DNNs rightfully raised concerns about their carbon footprint. The second part of the talk identifies the “4Ms” (Model, Machine, Mechanization, Map) that, if optimized, can reduce ML training energy by up to 100x and carbon emissions up to 1000x. By improving the 4Ms, ML held steady at <15% of Google’s total energy use despite it consuming ~75% of its floating point operations. Climate change is one of our most important problems, so ML papers should include emissions explicitly to foster competition on more than just model quality. External estimates have been off 100x–100,000x, so publishing emissions also ensures accurate accounting, which helps pinpoint the biggest challenges. With continuing focus on the 4Ms, we can realize the amazing potential of ML to positively impact many fields in a sustainable way.
David Patterson received BA, MS, and PhD degrees from UCLA. He is a UC Berkeley professor emeritus, a Google distinguished engineer, RIOS Laboratory Director, and the RISC-V International Vice-Chair. His most influential Berkeley projects were likely RISC and RAID. He received awards for teaching and service awards for his roles as ACM President, Berkeley CS Division Chair, and CRA Chair. The best known of his seven books is Computer Architecture: A Quantitative Approach. He and his co-author John Hennessy recently shared the 2017 ACM A.M Turing Award, the 2021 BBVA Foundation Frontiers of Knowledge Award, and the 2022 NAE Charles Stark Draper Prize for Engineering. The Turing Award is often referred to as the “Nobel Prize of Computing” and the Draper Prize is considered a “Nobel Prize of Engineering.”
Faculty Host: Ameet Talwalkar