Exploring code portability solutions for HEP with a particle tracking test code

TL;DR

This paper evaluates various code portability tools for high energy physics applications, focusing on their performance and implementation experience across different hardware architectures, including GPUs.

Contribution

It provides a comparative analysis of multiple portability solutions applied to a HEP tracking algorithm test code, highlighting their effectiveness and challenges.

Findings

Performance varies significantly across tools and architectures.

Some solutions offer easier implementation but lower efficiency.

The choice of portability method impacts scalability and maintainability.

Abstract

Traditionally, high energy physics (HEP) experiments have relied on x86 CPUs for the majority of their significant computing needs. As the field looks ahead to the next generation of experiments such as DUNE and the High-Luminosity LHC, the computing demands are expected to increase dramatically. To cope with this increase, it will be necessary to take advantage of all available computing resources, including GPUs from different vendors. A broad landscape of code portability tools -- including compiler pragma-based approaches, abstraction libraries, and other tools -- allow the same source code to run efficiently on multiple architectures. In this paper, we use a test code taken from a HEP tracking algorithm to compare the performance and experience of implementing different portability solutions.