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 looks at how to co-optimize power and performance for large-scale, accelerator-rich systems such as supercomputers and datacenters, especially in the face of fast-evolving workloads across high-performance computing (HPC) and Machine Learning (ML).

Specifically, I have studied variability across GPUs in large-scale heterogenous systems, and developed solutions for harnessing variability at the job scheduler level to achieve better job performance and resource utilization.

In my work, we observed that one of the key sources of variability is power excursions, where the power spikes above the GPU power limit for short durations. Managing these excursions effectively is crucial for reducing variability and achieving good performance. In collaboration with AMD Research and Advanced Development (RAD), I developed hardware- and firmware-level solutions (in submission) that manages these excursions to reduce variability.

Overall, my research shows how characterizing, harnessing and managing variability as well as power excursions across the stack can overcome power-related challenges and lead to efficient hardware and systems for across HPC and AI!

news

Sep 2026	Presented a lightning talk on my research, "Variability and Power Management for Large-Scale, Accelerator-Rich Systems" at the ACM AI Leadership Summit's doctoral consortium in Atlanta, GA.
Jun 2026	Presented "Minos: Systematically Classifying Performance and Power Characteristics of GPU Workloads on HPC Clusters" at ACM SIGMETRICS 2026 in Ann Arbor, MI. Presented "CRISPIER: A CRISP-Based DVFS Implementation in gem5 for AMD GPUs" at the gem5 Workshop at ISCA 2026, Raleigh, NC. I got selected for the doctoral consortium (Future Leaders of AI) at the inaugural ACM AI Leadership Summit 2026, to be held in Atlanta, GA.
May 2026	Workshop paper on modeling DVFS in gem5 GPU (with graduate student mentee Nishant Aggarwal) accepted at the 7th gem5 Users' Workshop!
Mar 2026	Our paper on GPU performance and power co-characterization — Minos — is accepted at SIGMETRICS 2026!
Dec 2025	Our paper (with MLCommons' Science Working Group) on AI Benchmark Carpentry posted on arXiv! Received AMD spotlight award (Q4 2025) for technical contributions to power estimation and excursion management.

selected publications

Minos: Systematically Classifying Performance and Power Characteristics of GPU Workloads on HPC Clusters

Rutwik Jain, Yiwei Jiang, Matthew D. Sinclair, and 1 more author

Proc. ACM Meas. Anal. Comput. Syst., May 2026

DOI Bib

@article{Minos-JainJiang,
  author = {Jain, Rutwik and Jiang, Yiwei and Sinclair, Matthew D. and Venkataraman, Shivaram},
  title = {Minos: Systematically Classifying Performance and Power Characteristics of GPU Workloads on HPC Clusters},
  year = {2026},
  issue_date = {June 2026},
  publisher = {Association for Computing Machinery},
  address = {New York, NY, USA},
  volume = {10},
  number = {2},
  url = {https://doi.org/10.1145/3805644},
  doi = {10.1145/3805644},
  journal = {Proc. ACM Meas. Anal. Comput. Syst.},
  month = may,
  articleno = {46},
  numpages = {27},
  keywords = {classification, gpgpu, hpc, power, performance},
  bibtex_show = true,
  selected = true
}

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
}