Conference Presentation (After Call)

FZJ-2023-02331

http://join2-wiki.gsi.de/foswiki/pub/Main/Artwork/join2_logo100x88.png Continuous Integration for HPC

Fritz, J. (Corresponding author)FZJ* ; Kabadshow, I.FZJ* ; Speck, R.FZJ*

2023

Please use a persistent id in citations: http://hdl.handle.net/2128/34549

Abstract: This project-talk shall give an update on the current status of the CI-HPC project within JLESC. In the last JLESC-meeting some issues and aspects of CI-HPC were raised, that have been taken care of. Two shall be presented here.First, an approach to combine best of both worlds from GitHub and GitLab: The large community and visibility of GitHub with the rich feature set that is available in the CI of Gitlab. This is especially relevant if there are self-hosted Gitlab-Instances with access to computing infrastructure, that is not reachable from the outside otherwise.Another aspect was how to simplify the setup of Gitlab-Runners. As the architecture of HPC-Systems changes, the testing of the code that runs on those machines should also change. Therefore, it is important to execute developed code on machines with the same characteristics (e.g. architecture). But not all architectures offer the possibility to run docker-containers. For those cases SSH-Executors can be used in Gitlab-Runners to execute automated tests on remote machines. The talk will introduce a way to also setup this kind of runner easily. This makes it possible to run CI-Jobs on much more machines and on more architectures.

---

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. JLESC - Joint Laboratory for Extreme Scale Computing (JLESC-20150708) (JLESC-20150708)
3. RGRSE - RG Research Software Engineering for HPC (RG RSE) (RG-RSE) (RG-RSE)

Appears in the scientific report 2023

---

Database coverage:

---