Contribution to a conference proceedings/Contribution to a book FZJ-2025-05251

http://join2-wiki.gsi.de/foswiki/pub/Main/Artwork/join2_logo100x88.png
RSEHPC@ISC25: Tools and Techniques for Continuous Integration and Benchmarking

 ;  ;

2026
Springer Nature Switzerland Cham
ISBN: 978-3-032-07611-3 (print), 978-3-032-07612-0 (electronic)

High Performance Computing
ISC High Performance 2025, ISC25, HamburgHamburg, Germany, 22 Jun 2025 - 26 Jun 20252025-06-222025-06-26
Cham : Springer Nature Switzerland, Lecture Notes in Computer Science 16091, 683 - 692 () [10.1007/978-3-032-07612-0_54]

This record in other databases:  

Please use a persistent id in citations: doi:

Abstract: This summary paper presents the outcomes of the RSEHPC workshop at ISC25, which convened experts from the Research Software Engineering and High-Performance Computing communities to examine the integration of Continuous Integration and Continuous Benchmarking (CI/CB) throughout the HPC system life cycle. Through keynote and lightning talks as well as through lively discussions, participants explored the use of CI/CB from system design and procurement to deployment and production use. The paper highlights three case studies: KIT’s CI/CB services in production for flexible HPC benchmarking; JSC’s deployment of exaCB for system-wide benchmarking during the JUPITER exascale system preparations; and RIKEN R-CCS’s integration of CI/CB in the early design phase of the FugakuNEXT project. User perspectives from software developers of projects like OpenFOAM, PIConGPU, and libNEGF underscore both the growing importance and the practical challenges of implementing CI/CB in diverse and complex HPC environments. A recurring theme is the tension between the need for customized solutions and the drawbacks of fragmentation, which hinder reproducibility, portability, and collaboration. The paper concludes with a call to action for the HPC community to establish common standards for CI/CB workflows and data formats. By making benchmarking practices more FAIR (Findable, Accessible, Interoperable, Reproducible), the community can enhance software sustainability and operational excellence in HPC ecosystems.


Contributing Institute(s):
  1. Jülich Supercomputing Center (JSC)
Research Program(s):
  1. 5112 - Cross-Domain Algorithms, Tools, Methods Labs (ATMLs) and Research Groups (POF4-511) (POF4-511)
  2. HiRSE - Helmholtz Platform for Research Software Engineering (HiRSE-20250220) (HiRSE-20250220)
  3. RGRSE - RG Research Software Engineering for HPC (RG RSE) (RG-RSE) (RG-RSE)

Database coverage:
NationallizenzNationallizenz ; SCOPUS
Click to display QR Code for this record

The record appears in these collections:
Document types > Events > Contributions to a conference proceedings
Document types > Books > Contribution to a book
Workflow collections > Public records
Institute Collections > JSC
Publications database

 Record created 2025-12-11, last modified 2025-12-21


Restricted:
Download fulltext PDF
Rate this document:

Rate this document:
1
2
3
 
(Not yet reviewed)