- This event has passed.
Colloquium – Muthu Baskaran
November 5, 2021 @ 1:00 pm - 2:00 pm
Dr. Muthu Baskaran
Friday, November 5 2021, 1:00pm Mountain
This will be a virtual talk over Zoom, but please feel free to join the in-person attendees (and refreshments) in the Utah Arch Lab (MEB 2170).
Title: R-Stream: An Automatic Optimizing and Parallelizing Compiler for HPC and Deep Learning
Abstract: The hardware architecture of modern supercomputers and exascale systems is being designed to include novel and/or specialized processors, resulting in increasingly heterogeneous and complex platforms. This shift demands advanced software solutions in the form of new programming models and runtimes, which in turn are complex, require verbose expressions of algorithms, and demand expensive and rare expertise. At Reservoir Labs, we develop an automatic optimizing and parallelizing polyhedral compiler called R-Stream that provides a high-utility, low-risk productive pathway for achieving portable performance on heterogeneous exascale platforms.
Our current work on R-Stream involves broadening and deepening the compiler support for automatic mapping to established and emerging programming models that are used for HPC and exascale computing. Our current work also involves extending R-Stream to integrate seamlessly with modern deep learning frameworks to efficiently map deep learning models onto novel and modern AI chips. In this talk, I’ll be presenting some background on R-Stream and some of the latest developments in R-Stream for mapping to HPC/exascale and deep learning systems.
Bio: Dr. Muthu Baskaran, Fellow and Managing Engineer at Reservoir Labs, is an expert in compilers and high-performance computing (HPC). He is one of the principal leads of Reservoir Labs’ R-Stream compiler. He has been developing advanced compiler techniques for modern high-performance parallel systems, exascale systems, and low-power embedded systems. He is the primary inventor of several patented/patent-pending techniques in R-Stream. His work has been published in several top-tier peer-reviewed conferences and journals. He has also served in the Technical and Program Committees of several top-tier HPC conferences and forums. He has won Best Paper Awards at IEEE HPEC Conference for his innovative work on high-performance optimizations.