Portable to Efficient: Auto-Tuning Hardware-Agnostic GPU Kernels in Julia

Proceedings - International Workshops on Parallel Processing, pp. 67–74

Abstract

Traditionally, GPU kernels have been developed and optimized within vendor-specific programming models to achieve high performance, resulting in software that is difficult to optimize and adapt across increasingly heterogeneous computing systems. Hardware-agnostic programming models offer a more sustainable approach to GPU software development by improving portability and maintainability, but achieving efficient execution across diverse architectures remains challenging. We address this challenge by integrating auto-tuning into hardware-agnostic GPU kernels written in Julia.

Authors 3

1. Floris-Jan Willemsen Aachen

   Leiden University

   Affiliation as printed

   LIACS, Leiden University, Leiden, Zuid-Holland, Netherlands

2. Evelyne Ringoot

   Massachusetts Institute of Technology

   Affiliation as printed

   Massachusetts Institute of Technology, Cambridge, Massachusetts, USA

3. Alan S. Edelman

   Massachusetts Institute of Technology

   Affiliation as printed

   Massachusetts Institute of Technology, Cambridge, Massachusetts, USA