Comparing OpenMP Implementations With Applications Across A64FX Platforms
Benjamin Michalowicz, Eric Raut, Yan Kang, Tony Curtis, Barbara, Chapman, Dossay Oryspayev

TL;DR
This paper compares the performance and scalability of OpenMP implementations of applications across A64FX platforms, including Fugaku and Ookami, highlighting differences due to compilers and hardware.
Contribution
It provides a comparative analysis of OpenMP application performance on A64FX-based supercomputers, focusing on scaling behavior and compiler effects.
Findings
OpenMP scaling varies significantly across applications.
Performance differences are influenced by compiler choices.
Applications achieve high scalability on A64FX platforms.
Abstract
The development of the A64FX processor by Fujitsu has created a massive innovation in High-Performance Computing and the birth of Fugaku: the current world's fastest supercomputer. A variety of tools are used to analyze the run-times and performances of several applications, and in particular, how these applications scale on the A64FX processor. We examine the performance and behavior of applications through OpenMP scaling and how their performance differs across different compilers on the new Ookami cluster at Stony Brook University as well as the Fugaku supercomputer at RIKEN in Japan.
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
