Krishna Inducted into HPCA Hall of Fame

With his eighth and ninth papers published in this year’s IEEE International Symposium on High-Performance Computer Architecture (HPCA), Tushar Krishna has been inducted into the organization’s Hall of Fame. Only 10 members of the HPCA community were inducted to the Hall of Fame this year — a significant distinction reserved for researchers with eight or more papers appearing in the proceedings of the symposium.

The 28th HPCA Symposium took place virtually April 2-6, 2022, where Krishna, an associate professor in Georgia Tech’s School of Electrical and Computer Engineering (ECE), was honored. The Symposium is hosted by IEEE (Institute of Electrical and Electronics Engineers) and provides a high-quality forum for scientists and engineers to present their latest research findings in the rapidly changing field.

Krishna had two papers at HPCA 2022. The first was titled “MAGMA: An Optimization Framework for Mapping Multiple DNNs on Multiple Accelerator Cores”. ECE Ph.D. student Sheng-Chun Kao co-authored the paper. It presented a software scheduler for efficiently mapping multiple Deep Neural Networks (DNN) on emerging hardware platforms that include multiple AI accelerators. An AI Accelerator refers to specialized hardware optimized for running AI workloads.

The second paper, “Stay in your Lane: A NoC with Low-overhead Multi-packet Bypassing” was co-authored with collaborators from University of Toronto, Texas A&M University, and University of Wisconsin-Madison. It presented a technique for packets stuck at routers in networks on chip (NoC) due to congestion or deadlocks to progress to their destination via high-priority bypass paths.

His research team in Synergy Lab at Georgia Tech won the Best Paper Award at the 26th HPCA Symposium in 2020. The team’s award-winning paper, "SIGMA: A Sparse and Irregular GEMM Accelerator with Flexible Interconnects for DNN Training,” showcased SIGMA, a flexible and scalable AI Accelerator that offers high utilization of all its processing elements for Deep learning (DL) — the premier algorithmic technique for analyzing data across multiple domains, especially in visual understanding, speech perception, and automated reasoning.

Krishna has been an ECE faculty member since 2015 with an adjunct appointment in the School of Computer Science. He held the ON Semiconductor Junior Professorship in ECE from 2019-2021. He serves as an associate director for the Center for Research into Novel Computing Hierarchies (CRNCH). Krishna’s research spans computer architecture, interconnection networks, networks-on-chip (NoC), and deep learning accelerators – with a focus on optimizing data movement in modern computing systems. His research is funded via multiple awards from NSF, DARPA, IARPA, Department of Energy, Intel, Google, Facebook, Qualcomm and TSMC. His papers have been cited over 11,000 times. Three of his papers have been selected for IEEE Micro’s Top Picks from Computer Architecture, one more received an honorable mention, and four have won best paper awards. He received the “Class of 1940 Course Survey Teaching Effectiveness” Award from Georgia Tech in 2018 and the “Roger P. Webb Outstanding Junior Faculty Award” from the School of ECE in Georgia Tech in 2021.