I am a PhD candidate in Computer Science at the University of Illinois at Urbana-Champaign, advised by Ravishankar K. Iyer. I am affiliated with the DEPEND research group at the Coordinated Science Laboratory. My research aims to improve the performance, resilience, and ease of management of large-scale heterogeneous computer systems.
I am particularly interested in applying ML to challenging problems within the design and management of computer systems. My dissertation research has demonstrated the application of domain-knowledge (i.e., knowledge about software and hardware architecture) as an inductive bias to ML techniques to solve a variety of systems problems, including heterogeneous resource management, distributed failure detection, and resilience assessment.
- Sep 5, 2020 Our SC 2020 paper has been nominated for the best paper and best student paper awards.
- Aug 15, 2020 Our paper on fine-grained ML-based resource rebalancing to meet microservice SLO-requirements has been accepted at OSDI 2020.
- Jul 3, 2020 Our paper on ML-based load-balancing in the Linux Kernel’s CFS has been accepted at ApSys 2020.
- Jun 29, 2020 Our paper on ML-based diagnosis and localization of large-scale distribted file-system failures has been accepted at Supercomputing 2020.
- May 31, 2020 Our paper on domain-knowledge guided reinforcement learning for scheduling tasks across CPUs, GPUs, and FPGAs has been accepted at ICML 2020.
Selected Publications [Full List: Publications, Projects]
Live Forensics for HPC Systems: A Case Study on Distributed Storage Systems.
- Best Paper Candidate
- Best Student Paper Candidate
ASAP: Accelerated Short Read Alignment on Programmable Hardware.
IEEE Transactions on Computers.