Adarsh Yoga

About

I am a compiler engineer in the Intel Python team. I work on designing programming abstractions for expressing GPU parallelism in Python. Before joining Intel, I went through the Santosh Nagarakatte Ph.D grind at Rutgers University. I worked on program analysis techniques for identifying performance and correctness issues in parallel programs.

Publications

• On-the-fly Data Race Detection with the Enhanced OpenMP Series-Parallel Graph
  Nader Boushehrinejadmoradi, Adarsh Yoga, and Santosh Nagarakatte
  International Workshop on OpenMP(IWOMP) 2020


• Paralleism-Centric What-If and Differential Analyses
  Adarsh Yoga, and Santosh Nagarakatte
  Programming Language Design and Implementation(PLDI) 2019
  [pdf] [Source]


• A Parallelism Profiler with What-If Analyses for OpenMP Programs
  Nader Boushehrinejadmoradi, Adarsh Yoga, and Santosh Nagarakatte
  International Conference for High Performance Computing, Networking, Storage, and Analysis(SC) 2018
  [pdf] [Source]


• A Fast Causal Profiler for Task Parallel Programs
  Adarsh Yoga,and Santosh Nagarakatte
  The Eurpoean Software Engineering Conference And The Foundations Of Software Engineering(ESEC/FSE) 2017
  [pdf] [Source]


• Path-Synchronous Performance Monitoring in HPC Interconnection Networks with Source-Code Attribution
  Adarsh Yoga,and Milind Chabbi
  International Workshop on Performance Modeling, Benchmarking and Simulation of High Performance Computer Systems(PBMS) 2017
  [pdf] [Source]


• Parallel Data Race Detection for Task Parallel Programs with Locks
  Adarsh Yoga, and Santosh Nagarakatte
  The Foundations Of Software Engineering(FSE) 2016
  [pdf] [Source]


• Atomicity Violation Checker for Task Parallel Programs
  Adarsh Yoga, and Santosh Nagarakatte
  International Symposium on Code Generation and Optimization(CGO) 2016
  [pdf] [Source]


• MATLAB Parallelization through Scalarization
  Chun-Yu Shei, Adarsh Yoga, Madhav Ramesh, and Arun Chauhan
  Workshop on Interaction between Compilers and Computer Architectures(INTERACT) 2011
  [Article DOI]

Patents

• Path-synchronous Performance Monitoring of Interconnection Networks based on Source Code Attribution
  Milind M Chabbi, Michael Schlansker, and Adarsh Yoga
  Patent No.: US10887213B2

PhD Dissertation

• Parallelism-Driven Performance Analysis Techniques for Task Parallel Programs
  PhD Dissertation, Rutgers University, October 2019
  [pdf]

Awards and Honors

• Bronze medal at the Student Research Competition at PLDI 2015
• Best Emerging Research Poster Award at the IEEE 2010 International Conference on Cloud Computing Technology and Science

Professional Activities

• Artifact Evaluation Committee member at PLDI 2016, 2020, and CGO-PPoPP 2017, 2019
• Student Volunteer at PLDI 2015, 2016, and 2018