| Title: |
What Will the Grace Hopper-Powered Jupiter Supercomputer Bring for Sparse Linear Algebra? |
| Authors: |
Tsai, Yu-Hsiang; Bode, Mathis; Anzt, Hartwig |
| Source: |
New York, NY, USA : ACM New York, NY, USA 228 - 235 (2026). doi:10.1145/3773656.3773691 ; SCA/HPCAsia 2026: Supercomputing Asia and International Conference on High Performance Computing in Asia Pacific Region, Osaka, Japan, 2026-01-26 - 2026-01-29 |
| Publisher Information: |
ACM New York, NY, USA |
| Publication Year: |
2026 |
| Collection: |
Forschungszentrum Jülich: JuSER (Juelich Shared Electronic Resources) |
| Subject Geographic: |
DE |
| Description: |
The first exascale supercomputer in Europe, JUPITER, is currently being built using the NVIDIA Grace Hopper superchips as main building blocks. JUPITER is designed to provide computing power for both data-driven (AI) workloads and numerics-based simulation workloads. For both workload types, and particularly for PDE-based simulations, high-performance sparse linear algebra operations are crucial. In this paper, we analyze the performance levels that sparse linear algebra operations can achieve on the JUPITER supercomputer and identify algorithmic modifications that can improve performance by acknowledging the Grace Hopper architecture. |
| Document Type: |
conference object |
| Language: |
English |
| Relation: |
info:eu-repo/grantAgreement/EC//101118139 |
| Availability: |
https://juser.fz-juelich.de/record/1052213; https://juser.fz-juelich.de/search?p=id:%22FZJ-2026-00840%22 |
| Rights: |
info:eu-repo/semantics/openAccess |
| Accession Number: |
edsbas.630AD2C3 |
| Database: |
BASE |