Rutwik Jain

7575 Morgridge Hall,

1209 University Ave,

Madison, WI 53715

I am a Ph.D. candidate in the Computer Sciences department at University of Wisconsin-Madison, advised by Prof. Matt Sinclair and Prof. Shivaram Venkataraman.

My research interests lie in the area of computer architecture and systems. Specifically, I have studied performance variability across GPUs in large-scale heterogenous systems, and developed solutions for harnessing variability at the job scheduler level to achieve better resource utilization.

Power management mechanisms, which make different decisions on different GPUs, are a key source of variability. Recently, I have been collaborating with AMD Research and Advanced Development (RAD) to look at cross-stack solutions that can effectively manage power spikes efficiently, improve performance and mitigate variability, while maintaining chip integrity.

news

Jun 2025	Graduate mentee Yiwei Jiang gave a talk about our work at AMD Research and Advanced Development
Jan 2025	I presented our work on GPU variability and variability-aware scheduling at AMD Matt and I presented our work on GPU variability to the MLCommons Science Working Group
Nov 2024	I presented PAL, which designs performance variability-aware schedulers for GPU-rich clusters, at SC’24 in Atlanta, GA.
Jul 2024	Our paper on designing variability-aware schedulers for GPU-rich clusters was accepted at SC’24.
May 2024	Started internship with AMD Research working with Srilatha (Bobbie) Manne’s team on power management for Instinct GPU products Matt presented on our work on GPU variability at NVIDIA

selected publications

PAL: A Variability-Aware Policy for Scheduling ML Workloads in GPU Clusters

Rutwik Jain, Brandon Tran, Keting Chen, and 2 more authors

In Proceedings of the International Conference for High Performance Computing, Networking, Storage, and Analysis, Atlanta, GA, USA, 2024

DOI Bib PDF

@inproceedings{Jain-SC24-PAL,
  author = {Jain, Rutwik and Tran, Brandon and Chen, Keting and Sinclair, Matthew D. and Venkataraman, Shivaram},
  title = {PAL: A Variability-Aware Policy for Scheduling ML Workloads in GPU Clusters},
  year = {2024},
  isbn = {9798350352917},
  publisher = {IEEE Press},
  url = {https://doi.org/10.1109/SC41406.2024.00032},
  doi = {10.1109/SC41406.2024.00032},
  booktitle = {Proceedings of the International Conference for High Performance Computing, Networking, Storage, and Analysis},
  articleno = {26},
  numpages = {18},
  keywords = {Cluster Scheduling, GPGPU, Machine Learning, Performance Variability, Power Management},
  location = {Atlanta, GA, USA},
  bibtex_show = true,
  selected = true,
  series = {SC '24}
}

Not All GPUs Are Created Equal: Characterizing Variability in Large-Scale, Accelerator-Rich Systems

Prasoon Sinha, Akhil Guliani, Rutwik Jain, and 3 more authors

In SC22: International Conference for High Performance Computing, Networking, Storage and Analysis , Nov 2022

DOI Bib

@inproceedings{Sinha-SC22-GPUVar,
  author = {Sinha, Prasoon and Guliani, Akhil and Jain, Rutwik and Tran, Brandon and Sinclair, Matthew D. and Venkataraman, Shivaram},
  booktitle = { SC22: International Conference for High Performance Computing, Networking, Storage and Analysis },
  title = {{ Not All GPUs Are Created Equal: Characterizing Variability in Large-Scale, Accelerator-Rich Systems }},
  year = {2022},
  volume = {},
  issn = {},
  pages = {1-15},
  doi = {10.1109/SC41404.2022.00070},
  url = {https://doi.ieeecomputersociety.org/10.1109/SC41404.2022.00070},
  publisher = {IEEE Computer Society},
  address = {Los Alamitos, CA, USA},
  bibtex_show = true,
  selected = true,
  month = nov
}