Performance Versus Maintainability: A Case Study of Scream on Frontier

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

The Simple Cloud-Resolving E3SM Atmosphere Model (Scream) won the inaugural ACM Gordon Bell Prize for Climate Modeling. While most of Scream is portable Kokkos code, the Gordon-Bell runs did include tuning specifically for Frontier, the exascale computer at Oak Ridge National Laboratory. Production science runs use the same high-level configuration of Scream, but the tuned kernels do not meet the software standards necessary to merge into the production code base. This work describes experiments to refactor these kernels to meet the maintainability requirements of the production Scream code base while preserving high performance.

Original languageEnglish
Title of host publication2024 IEEE International Parallel and Distributed Processing Symposium Workshops, IPDPSW 2024
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages286-292
Number of pages7
ISBN (Electronic)9798350364606
DOIs
StatePublished - 2024
Event2024 IEEE International Parallel and Distributed Processing Symposium Workshops, IPDPSW 2024 - San Francisco, United States
Duration: May 27 2024May 31 2024

Publication series

Name2024 IEEE International Parallel and Distributed Processing Symposium Workshops, IPDPSW 2024

Conference

Conference2024 IEEE International Parallel and Distributed Processing Symposium Workshops, IPDPSW 2024
Country/TerritoryUnited States
CitySan Francisco
Period05/27/2405/31/24

Keywords

  • GPUs
  • Kokkos
  • performance;
  • software maintainability

Fingerprint

Dive into the research topics of 'Performance Versus Maintainability: A Case Study of Scream on Frontier'. Together they form a unique fingerprint.

Cite this