LLVM in HPC: Enabling Performance Portability, Interoperability, and Novel Features - Part 2

Explore recent advancements in the LLVM compiler framework for high-performance computing (HPC) in this seminar talk by Dr. Johannes Doerfert from Lawrence Livermore National Laboratory. Delve into topics such as portable CUDA, debugging and tuning at scale, effortless GPU execution of legacy codes, automatic differentiation of HPC programming models, machine learning in compilers, and the impact of missing static information. Gain insights into how these developments aim to enhance performance, tooling, and development in HPC, with a focus on integrating improvements into the LLVM community compiler for widespread impact across various toolchains and programming languages used in scientific computing.

LLVM in HPC (part 2): Enabling Performance Portability, Interoperability, and Novel Features