13th CÉCI Users Meeting

Thursday 18 Apr 2024, 12:45 → 18:00 Europe/Brussels

CYCL01 (Bâtiment de Hemptine - Louvain-la-Neuve)

                                                   

Groupe de Contact FNRS "Calcul Intensif" - Louvain-la-Neuve, April 18th, 12:45 - 18:00

This meeting gathers all the CÉCI users and their PIs from the five member universities. It is a nice opportunity to discover how other users are (ab)using the clusters, learn about new supercomputers available to you, and meet in person and share ideas with the CÉCI board and the system administrators.

 

Provisional schedule:

12h45: Doors open
13h00-13h05: Welcome speach by Vice Rector Prof. Michel Verleysen
13h05-14h05: Keynote by Paul Fischer (world leader in computational fluid dynamics)
14h05-15h00: CÉCI Technical talks
15h00-15h15: Coffee break
15h15-15h45: Tier-0 and Tier-1 technical talk
15h45-17h00: Scientific talks
17h: Drink
 

1 Welcome

Speaker: Michel Verleysen

2 Exascale Computing: A User’s Perspective

We discuss design and performance issues for computational science applications on GPU-based HPC platforms from a user/developer perspective. We focus particularly on the strong-scale limit, where HPC users typically run in practice. This limit is closely tied to the local problem size, that is, the number of degrees-of-freedom (or grid points) per MPI rank, which reflects the amount of parallel work for domain-decomposition-based parallelism. As this number decreases, communication effects become important. On GPUs, kernel launch latency is another factor controlling achievable speed-ups. We discuss several features that distinguish algorithmic optimization strategies for GPU- based HPC platforms from their predecessor, fine-grained distributed-memory, platforms such as IBM’s BG series. As concrete examples, we consider the performance of high-order spectral element methods for incompressible flow on all of Mira ( > 1 M ranks), all of Summit (> 27,000 ranks), and all of Frontier (> 70,000 ranks). We discuss optimizations for each of these platforms, with a particular focus on the Poisson problem, which is the stiffest and therefore most time-consuming substep in Navier-Stokes time advancement. Examples are presented in the context of Nek5000/RS, which is a high-order open-source code for thermal-fluid transport problems.

---

Paul Fischer received his Ph.D. in mechanical engineering from MIT in 1989.After postdoctoral research at MIT and Caltech, he joined Brown University as an assistant professor in the Applied Mathematics department. In 1998, he was hired as a mathematician at Argonne and was promoted to senior scientist in 2008. He is a world leader in computational fluid dynamicis.

More information about Paul Fisher on the Aragone National Laboratory website: https://www.anl.gov/profile/paul-f-fischer

Speaker: Paul Fischer (Aragone National Laboratory)

BELGIUM1s.pdf

3 CÉCI technical talk

Topics:

• Lemaitre4
• Lyra
• Common storage (CEPH)
• Hercules and Dragon
• Help us help you

CÉCI technical talk.pdf

helpushelpyou.pdf

15:00 Coffee break

4 Tier-1 and Tier-0 technical talk

Speaker: Mr Orian Louant (Université de Liège)

lucia-lumi-eurohpc-ceci-user-meeting-2024.pdf

5 Population genomic analysis using CÉCI clusters

Speaker: Surabhi Ranavat

6 Scaling self-supervised pre-training on Lucia and Lumi

Speaker: Renaud Vandeghen (ULiège)

7 Molecular Dynamics Simulations of multimillion-atom systems using the power of Lucia and Lumi GPUs

Speaker: David Dellemme

8 Integrating AI-Driven heterogenous HPC for Advanced Optical Design

Speaker: Nicolas Roy (UNamur)

9 An unbounded Multigrid Poisson solver within an adaptive multiresolution framework for the Navier-Stokes equations, MURPHY

Speaker: Gilles Poncelet (UCLouvain)

17:00 Drink