Vijay Thakkar

I’m a graduate student in the College of Computing at Georgia Institute of Technology, advised by Dr. Rich Vuduc.

I work on high performance computing, focusing primarily on accelerated computing and programming models for heterogeneous architectures. My research helps HPC developers write speed of light applications and kernels that take advantage of modern hardware capabilities in a way that does not make them want to pull their hair out.

My interests mostly lie at the intersection of hardware and software for accelerated computing. I have worked at various hardware shops to this end including Arm, Cerebras Systems, and Nvidia on hardware modeling, high performance kernels, and library design. Good folks at Oak Ridge national lab’s OLCF remain my long time collaborators on application level projects.

Currently, I work at NVIDIA full time while I finish my PhD. At NVIDIA, I collaborate closely with Cris Cecka from NVR to lead the design of next generation linear algebra libraries, namely, CUTLASS 3.0, a project I have been working on since its inception. I also work on the exposure of next generation tensor core hardware features into the CUDA C++ programming model for peak developer productivity without performance compromises.

News

[Subscribe]

Publications

[DBLP]

[Google Scholar]

Conference Papers

• Exaflops Biomedical Knowledge Graph Analytics.

  Ramakrishnan Kannan and Piyush Sao and Hao Lu and Jakub Kurzak and Gundolf Schenk and Yongmei Shi and Seung-Hwan Lim and Sharat Israni and Vijay Thakkar and Guojing Cong and Robert Patton and Sergio E. Baranzini and Richard Vuduc and Thomas Potok.

  Supercomputing 2022. ACM Gordon Bell award finalist.

  • Paper (PDF)

• Scalable all-pairs shortest paths for huge graphs on multi-GPU clusters.

  Piyush Sao, Hao Lu, Ramakrishnan Kannan, Vijay Thakkar, Richard Vuduc, and Thomas Potok.

  HPDC 2021.

  • Paper (PDF)

• Scalable knowledge graph analytics at 136 petaflop/s.

  Ramakrishnan Kannan, Piyush Sao, Hao Lu, Drahomira Herrmannova, Vijay Thakkar, Robert Patton, Richard Vuduc, and Thomas Potok.

  Supercomputing 2020. ACM Gordon Bell award finalist.

  • Paper (PDF)

• Conditioning deep generative raw audio models for structured automatic music.

  Vijay Thakkar, Rachel Manzelli, Ali Siahkamari, and Brian Kulis.

  ISMIR 2018.

  • Paper (PDF)

Workshop Papers

• An end to end model for automatic music generation: Combining deep raw and symbolic audio networks.

  Rachel Manzelli, Vijay Thakkar, Ali Siahkamari, and Brian Kulis.

  MuME 2018.

  • Paper (PDF)