Michael Bauer

Michael Bauer joined NVIDIA Research in October of 2014 after finishing his PhD in computer science at Stanford University. As part of his thesis, he developed the Legion programming system for high performance supercomputing codes. Legion is currently deployed in several production applications running on the top supercomputers in the world. Michael was also the primary author of both the CudaDMA library and Singe DSL compiler, both of which were early examples of warp-specialized GPU programming.

Research Area(s)

High Performance Computing

Programming Languages, Systems and Tools

Publications

2025

Task-Based Tensor Computations on Modern GPUs

Rohan Yadav, Michael Garland, Alex Aiken, Michael Bauer

PLDI

Composing Distributed Computations Through Task and Kernel Fusion

Rohan Yadav, Shiv Sundrum, Wonchan Lee, Michael Garland, Michael Bauer, Alex Aiken, Fredrik Kjolstad

ASPLOS

Automatic Tracing in Task-Based Runtime Systems

Rohan Yadav, Michael Bauer, David Broman, Michael Garland, Alex Aiken, Fredrik Kjolstad

ASPLOS

2023

Legate Sparse: Distributed Sparse Computing in Python

Rohan Yadav, Wonchan Lee, Melih Elibol, Taylor Patti, Manolis Papadakis, Michael Garland, Alex Aiken, Fredrik Kjolstad, Michael Bauer

Supercomputing

Visibility Algorithms for Dynamic Dependence Analysis and Distributed Coherence

Michael Bauer, Elliott Slaughter, Sean Treichler, Wonchan Lee, Michael Garland, Alex Aiken

PPoPP

2021

Scaling Implicit Parallelism via Dynamic Control Replication

Michael Bauer, Wonchan Lee, Elliott Slaughter, Zhihao Jia, Mario Di Renzo, Manolis Papadakis, Galen Shipman, Patrick McCormick, Michael Garland, Alex Aiken

Principles and Practices of Parallel Programming (PPoPP)