Pedro Bruel

Performance Measurement and Tuning

I obtained my PhD on Computer Science from the Universities of Grenoble Alpes, France, and of São Paulo, Brazil. My advisors were Alfredo Goldman and Arnaud Legrand, and I worked with the POLARIS Inria Team at LIG and the CCSL at the University of São Paulo.

My research interests are automatic performance measurement and tuning for High Performance Computing, using machine learning, search heuristics, and design of experiments.

My PhD defense was recorded, and you can also check the complete thesis, the slides, and the git repository.

news

Jul 10, 2021 I defended my PhD! Here’s the recording
May 27, 2021 Thesis defense scheduled for July 9th! Will post streaming link when it is available.

selected publications

  1. Autotuning under Tight Budget Constraints: A Transparent Design of Experiments Approach
    Bruel, Pedro, Quinito Masnada, Steven, Videau, Brice, Legrand, Arnaud, Vincent, Jean-Marc, and Goldman, Alfredo
    In The 19th Annual IEEE/ACM International Symposium in Cluster, Cloud, and Grid Computing (CCGrid 2019) 2019
  2. Autotuning CUDA compiler parameters for heterogeneous applications using the OpenTuner framework
    Bruel, Pedro, Amarís, Marcos, and Goldman, Alfredo
    Concurrency and Computation: Practice and Experience 2017