@@ -37,6 +37,55 @@ could it have helped? And what did hinder the adaptation of CI/CB?
3737
3838The RSEHPC@ISC25 will be held in-person on 13th June 2025 in Hamburg.
3939
40+ ### Keynotes
41+
42+ #### Cx in production
43+
44+ Speaker: [ Michele Mesiti] ( https://www.scc.kit.edu/en/staff/16297.php ) , Scientific Computing Centre, Karlsruhe Institute of Technology, Germany
45+
46+ Abstract:
47+ TBA
48+
49+ #### The Impact of CI/CD/CB during JUPITER’s Ascension
50+
51+
52+ Speaker: [ Jayesh Badwaik] ( https://www.fz-juelich.de/profile/badwaik_j ) , Jülich Supercomputing Centre, Forschungszentrum Jülich GmbH, Germany
53+
54+ Abstract:
55+ Ensuring performance consistency and early regression detection is critical in
56+ high-performance computing (HPC) operations. Traditional benchmarking methods
57+ rely on manual execution, leading to inconsistencies and delayed issue
58+ detection. During the JUREAP program, we integrated Continuous Benchmarking
59+ (CB) using exacb to standardize performance evaluation across 50 applications.
60+ This automation improved reproducibility, streamlined reporting, and enabled
61+ early detection of system anomalies, such as faulty Slurm updates and workflow
62+ execution issues on the JEDI machine. Even without a fully operational
63+ exascale supercomputer, exacb facilitated systematic performance comparisons,
64+ providing valuable insights into application scalability.
65+
66+ Beyond JUREAP, CI/CD/CB enhances research software development and HPC system
67+ management. Our framework simplifies benchmarking, ensuring efficient performance
68+ tracking and optimization at scale—key for the upcoming JUPITER exascale
69+ supercomputer. Automating benchmarking reduces manual overhead, improves
70+ system stability, and aids in troubleshooting by providing structured
71+ performance insights. In this talk, we share our experience implementing CB in
72+ JUREAP, key findings from benchmarking 50 applications, and the broader impact
73+ of CI/CD/CB on research software, system administration, and future exascale
74+ computing.
75+
76+ #### Utilization of Cx in Fugaku and FugakuNEXT
77+
78+ Speaker: [ Hitoshi Murai] ( https://www.r-ccs.riken.jp/en/research/octd/sdt/ ) , RIKEN Center for Computational Science, Japan
79+
80+ Abstract:
81+ RIKEN R-CCS starts the FugakuNEXT project from April, 2025. FugakuNEXT is a
82+ project for developing Japanese next-generation supercomputer. Unlike its
83+ predecessor, Supercomputer Fugaku, it is equipped with GPUs and targets both HPC
84+ and AI workloads; therefore it is crucial to develop and/or optimize applications
85+ for such a system. For this goal, we are planning to utilize Cx technologies.
86+ In this talk, we will explain briefly our plan of utilizing Cx in the FugakuNEXT
87+ project along with that in Fugaku.
88+
4089### Contribution submissions:
4190To present a lightning talks at RSEHPC@ISC25 use this [ submission form] ( https://forms.gle/4QP9GNPU2jncTuHa9 ) .
4291RSEHPC@ISC25 will emphasise an open and inclusive atmosphere and we encourage proposals from a diverse range of areas and backgrounds.
0 commit comments