Comparing the behavior of OpenMP Implementations with various Applications on two different Fujitsu A64FX platforms

TL;DR

This paper analyzes the performance and behavior of various OpenMP applications on two Fujitsu A64FX platforms, highlighting differences in scalability and implementation effects across different hardware and compiler configurations.

Contribution

It provides a comparative analysis of OpenMP application performance on two Fujitsu A64FX systems, revealing insights into scalability and implementation impacts.

Findings

Performance varies across applications and platforms.

Compiler choices influence scalability and efficiency.

Hardware differences affect application behavior.

Abstract

The development of the A64FX processor by Fujitsu has been a massive innovation in vectorized processors and led to Fugaku: the current world's fastest supercomputer. We use a variety of tools to analyze the behavior and performance of several OpenMP applications with different compilers, and how these applications scale on the different A64FX processors on clusters at Stony Brook University and RIKEN.