Tapan Chugh

pic.jpg

I am a PhD student at the University of Washington, where I am advised by Arvind Krishnamurthy and Ratul Mahajan. I am broadly interested in distributed systems and machine learning systems. My thesis research revolves around designing performance models and algorithms to optimize resource efficiency in large-scale machine learning deployments. My ongoing projects seek to improve the quality and efficiency of long context generations in large language models.

My academic journey has been complemented by valuable practical experiences: I am actively contributing to the Model Context Protocol specification and open-source projects in that ecosystem. Until June 2025, I was a student researcher at Systems Research @ Google. Previously, I collaborated with Srikanth Kandula and Ishai Menache as a research intern and visitor at Microsoft Research. Before starting my PhD, I spent two years as a Research Fellow at Microsoft Research India working with Muthian Sivathanu.

selected publications

Publications

  1. NSDI
    Efficient Direct-Connect Topologies for Collective Communications
    Liangyu Zhao, Siddharth Pal, Tapan Chugh, Weiyang Wang, Jason Fantl, Prithwish Basu, Joud Khoury, and Arvind Krishnamurthy
    In 22nd USENIX Symposium on Networked Systems Design and Implementation, NSDI 2025, Philadelphia, PA, USA, April 28-30, 2025, 2025
    PDF
  2. SoCC
    Anticipatory Resource Allocation for ML Training
    Tapan Chugh, Srikanth Kandula, Arvind Krishnamurthy, Ratul Mahajan, and Ishai Menache
    In Proceedings of the 14th ACM Symposium on Cloud Computing, 2023
    PDF
  3. SIGMETRICS
    Dremel: Adaptive Configuration Tuning of RocksDB KV-Store
    Chenxingyu Zhao, Tapan Chugh, Jaehong Min, Ming Liu, and Arvind Krishnamurthy
    Proceedings of the ACM on Measurement and Analysis of Computing Systems, 2022
    PDF
  4. SIGCOMM
    Gimbal: enabling multi-tenant storage disaggregation on SmartNIC JBOFs
    Jaehong Min, Ming Liu, Tapan Chugh, Chenxingyu Zhao, Andrew Wei, In Hwan Doh, and Arvind Krishnamurthy
    In Proceedings of the 2021 ACM SIGCOMM 2021 Conference, 2021
    PDF
  5. ASPLOS
    Astra: Exploiting predictability to optimize deep learning
    Muthian Sivathanu, Tapan Chugh, Sanjay S Singapuram, and Lidong Zhou
    In Proceedings of the Twenty-Fourth International Conference on Architectural Support for Programming Languages and Operating Systems, 2019
    PDF