Biography
Damien Lebrun-Grandié is a Senior Computational Scientist at Oak Ridge National Laboratory with over a decade of experience in the field. He holds a PhD in Nuclear Engineering from Texas A&M University, a MSc in Physics from the Karlsruhe Institute of Technology in Germany, and a MEng in Physics Engineering from Grenoble INP in France. His research focuses on developing algorithms and enabling technologies for solving large-scale, complex engineering, and scientific problems. As a founding member of the High Performance Software Foundation, Damien was instrumental in getting the organization started and continues to play a leading role on the Governing Board, representing the Technical Advisory Council. He is also the co-lead of the Kokkos C++ performance portability project, where he oversees a large international team of developers and researchers. Additionally, he represents ORNL on the C++ Standards Committee, where he has been a key contributor to foundational features for scientific computing like std::mdspan in C++23 and std::linalg for C++26.
SRP Project Title
Kokkos Tools
HPSF Project
Kokkos
Topical Areas
Applied Computer Science; High Performance Computing; Performance Evaluation and Benchmarking; Software Engineering
Abstract
This internship focuses on developing Kokkos Tools, a vital suite of libraries for analyzing and optimizing Kokkos applications without changing the source code. You will help expand the Kokkos Tools ecosystem by building: Performance & Energy Profiling tools to measure execution time, memory usage, and energy consumption. This work helps developers identify bottlenecks and optimize code for diverse hardware, including CPUs and GPUs. Code Sanity & Debugging tools to detect and diagnose common programming errors and invalid usage of the Kokkos API, ensuring code correctness. You will contribute to a major open-source project used globally in High-Performance Computing (HPC). You'll gain hands-on experience in parallel programming, performance-portability, and the full software development lifecycle. This is a chance to build critical skills in performance analysis and debugging while connecting with the international HPC community.
Desired Skills
We're looking for candidates with a passion for High-Performance Computing (HPC) and parallel programming. Relevant backgrounds could include Computer Science, Engineering, or a related technical field. Ideal interests and skills include: Experience with C++ (or a similar high-performance language). Familiarity with parallel programming models (e.g., CUDA, OpenMP, MPI, or Kokkos). An interest in performance analysis and finding ways to optimize code. A desire to contribute to a large-scale open-source project.
Lightning Talk Title
Your Summer Project: Taming the World's Fastest GPUs
Keywords
Performance Portability; Exascale Computing; C++; Heterogeneous Architectures; GPU; Parallel Programming; Tooling / Profiling