Tong Zhou

About Me

I received my PhD in Computer Science from Georgia Tech, working with the Habanero folks on compiler optimization, code generation and parallel computing. My research aims to reduce the complexity of writing HPC code via innovative compiler techniques such that domain scientists can use, e.g. Python, and still achieve high performance. In recent years, my research focus has been generating efficient CPU and GPU code for matrix-oriented programs written in high-level programming models. Please feel free to check out the following projects:

• VCSparse (Try Online): Generate fused and vectorizable kernels for sparse tensor programs in Python on CPUs and GPUs.
• APPy (Try Online): JIT-Compile Python loops and tensor expressions with simple annotations to run on GPUs.
• Intrepydd (Try Online): Compile Python programs from data science domains to C++ kernels for execution on CPUs.

Feel free to contact me (tz@gatech.edu) if you want to make your code (especially Python code) run faster 🚀. For anyone who's insterested in learning more about high performance computing, I highly recommend these materials 📚. Besides, I occasionally write technical blogs on Zhihu 💻.

My Curriculum Vitae is available here.

Publications

• High-Level Compiler Optimizations for Python Programs.
  Tong Zhou.
  PhD Dissertation. April 2024.
  [Abstract] [PDF] [Slides]

• APPy: Annotated Parallelism for Python on GPUs.
  Tong Zhou, Jun Shirako, Vivek Sarkar.
  Proceedings of the 33rd ACM SIGPLAN International Conference on Compiler Construction (CC 2024). March 2024.
  [Abstract] [PDF] [Slides] [GitHub] [Try It Online]

• ReACT: Redundancy-Aware Code Generation for Tensor Expressions.
  Tong Zhou, Ruiqin Tian, Rizwan A Ashraf, Roberto Gioiosa, Gokcen Kestor, Vivek Sarkar.
  Proceedings of the International Conference on Parallel Architectures and Compilation Techniques (PACT 2022). October 2022.
  [Abstract] [PDF] [Slides] [Demo]

• Intrepydd: Performance, Productivity, and Portability for Data Science Application Kernels.
  Tong Zhou, Jun Shirako, Anirudh Jain, Sriseshan Srikanth, Thomas M Conte, Richard Vuduc, Vivek Sarkar.
  Proceedings of the 2020 ACM SIGPLAN International Symposium on New Ideas, New Paradigms, and Reflections on Programming and Software (Onward! 2020). November 2020.
  [Abstract] [PDF] [Slides] [GitHub] [Try It Online]

• Valence: Variable Length Calling Context Encoding.
  Tong Zhou, Michael R. Jantz, Prasad A. Kulkarni, Kshitij A. Doshi, and Vivek Sarkar.
  28th ACM SIGPLAN International Conference on Compiler Construction (CC 2019). February 2019.
  [Abstract] [PDF] [Slides] [GitHub]

• MemBrain: Automated Application Guidance for Hybrid Memory Systems.
  M. Benjamin Olson, Tong Zhou, Michael R. Jantz, Kshitij A. Doshi, M. Graham Lopez, and Oscar Hernandez.
  13th IEEE International Conference on Networking, Architecture, and Storage (NAS 2018). October 2018. (Awarded Best Paper)
  [Abstract] [PDF]

• On Automated Feedback-Driven Data Placement in Multi-tiered Memory.
  T. Chad Effler, Adam P. Howard, Tong Zhou, Michael R. Jantz, Kshitij A. Doshi, Prasad A. Kulkarni.
  Proceedings of the 31st International Conference on Architecture of Computing Systems, ARCS 2018.
  [Abstract] [PDF]

Quotes of the Day

• "The hour of departure has arrived, and we go our separate ways, I to die, and you to live. Which of these two is better, only God knows." --- Socrates