IPDPS 2023 - I traveled to St. Petersburg Florida to present our new paper on Kernel Launcher at iWAPT2023, the International Workshop on Automatic Performance Tuning 2023 co-located with IPDPS. Kernel Launcher is a C++ library for creating optimal-performance portable CUDA applications.
Abstract
Graphic Processing Units (GPUs) have become ubiquitous in scientific computing. However, writing efficient GPU kernels can be challenging due to the need for careful code tuning. To automatically explore the kernel optimization space, several auto-tuning tools - like Kernel Tuner - have been proposed. Unfortunately, these existing auto-tuning tools often do not concern themselves with integration of tuning results back into applications, which puts a significant implementation and maintenance burden on application developers. In this work, we present Kernel Launcher: an easy-to-use C++ library that simplifies the creation of highly-tuned CUDA applications. With Kernel Launcher, programmers can capture kernel launches, tune the captured kernels for different setups, and integrate the tuning results back into applications using runtime compilation. To showcase the applicability of Kernel Launcher, we consider a real-world computational fluid dynamics code and tune its kernels for different GPUs, input domains, and precisions.
Citation
Kernel Launcher: C++ library for creating optimal-performance portable CUDA applications
S. Heldens, B. van Werkhoven
International Workshop on Automatic Performance Tuning (iWAPT2023) co-located with IPDPS 2023
https://doi.org/10.1109/IPDPSW59300.2023.00126
