Harini Muthukrishnan is a Research Scientist at NVIDIA’s Architecture Research Group since February 2022. Her work mainly focuses on GPU system architecture, virtualization, and scalability. Harini received a PhD from University of Michigan Computer Science and Engineering department in 2022. At Michigan, she was advised by Prof. Thomas Wenisch and her thesis focused on developing hardware and software solutions to improve multi-GPU scaling. Her research has been published at Top-tier architecture conferences such as ISCA, MICRO and HPCA, as well as recognized by a MICRO best paper nomination and an IEEE top picks honorable mention.

Interests
  • Computer Architecture
  • GPU Architecture
  • High Performance Computing
Education
  • PhD in Computer Science and Engineering, 2022

    University of Michigan

Recent Publications

FinePack: Transparently Improving the Efficiency of Fine-Grained Transfers in Multi-GPU Systems
GPS: A Global Publish-subscribe Model for Multi-GPU Memory Management
Efficient Multi-GPU Shared Memory Via Automatic Optimization of Fine-grained Transfers