Andreas HertenJun 12, 2024
3 GPU platforms and plenty of programming models – a while ago, I created a comparison table for the lab blog. For #SC23, I made it into a workshop paper for the #P3HPC workshop! I finally came around to write about it on the blog, from where you also can access the paper (which is non-open-access): https://doi.org/10.34732/xdvblg-0xiq5x
(I'm also requesting help from the community to keep everything up-to-date on GitHub: https://github.com/AndiH/gpu-lang-compat; #SC24 is coming up soon!) #HPC #GPU
Paper: Many Cores, Many Models – GPU Programming Model vs. Vendor Compatibility Overview

Various notes from the Accelerating Devices Lab (X-Dev) of Jülich Supercomputing Centre

JSC Accelerating Devices Lab
Caleb 🦈May 1, 2024

Finally! After half a year since #SC23, we finally have our very modestly sized #almalinux cluster!

CC: @jonathanspw @benny

Leibniz Supercomputing CentreMar 26, 2024

Es ist wieder #Quartl-Zeit! In der aktuellen Ausgabe des Mitteilungsblatts von KONWIHR und BGCE geht es u. a. um #SuperMUC-NG Phase 2, die Erweiterung von Simulationen mit KI-Methoden und es wird auf die #SC23 zurückgeblickt. #HPC #IT4Science @tu_muenchen

https://www.cs.cit.tum.de/fileadmin/w00cfj/sccs/Quartl/98/Quartl_109.pdf

HPC GuruFeb 15, 2024

Today Nvidia released a video that offers the first public look at #Eos, its latest data-center-scale #supercomputer

#Eos features a total of 4,608 #H100 #GPUs

#Eos was ranked #9 on the latest #Top500 list revealed at #SC23

https://blogs.nvidia.com/blog/eos/

#AI #HPC #GPU #DGX

NVIDIA Eos Revealed: Peek Into Operations of a Top 10 Supercomputer | NVIDIA Blog

Providing a peek at the architecture powering advanced AI factories, NVIDIA Thursday released a video that offers the first public look at Eos, its latest data-center-scale supercomputer. An extremely large-scale NVIDIA DGX SuperPOD, Eos is where NVIDIA developers create their AI breakthroughs using accelerated computing infrastructure and fully optimized software. Eos is built with 576 Read Article

NVIDIA Blog
ax3lFeb 13, 2024

Uploaded a new in situ visualization that we showed at #SC23 #ISAV23. Enjoy!

#ParticleAccelerator #HPC #GPU #WarpX #InSitu #visualization
https://www.youtube.com/watch?v=L75CfsiHJRg

Staged Laser Wakefield Acceleration (9 Stages, WarpX @SC23/ISAV23)

YouTube
Netherlands eScience CenterJan 22, 2024

New eScience Center blog post released!

RSE Alessio Sclocco wrote a blog about how he, Ben van Werkhoven, Stijn Heldens, and Floris-Jan Willemsen attended and presented a tutorial on energy-efficient GPU computing at the largest High-Performance Computing (HPC) conference in the world, which took place in Denver, USA.

This blog post covers some of the highlights of the conference and goes in deep into the content of the tutorial.⬇️

https://blog.esciencecenter.nl/talking-about-energy-efficient-gpu-computing-at-sc23-c85d75fd3928

#GPU #Computing #HPC #eScience #SC23

HPC GuruDec 14, 2023

With #exascale #supercomputers now a reality, the focus has shifted to optimizing energy consumption to ensure the cost of powering the system over its lifetime does not exceed the cost of the system itself

https://semiengineering.com/the-journey-to-exascale-computing-and-beyond/

#HPC #AI #SC23

The Journey To Exascale Computing And Beyond

Optimizing power consumption in the world's most powerful supercomputers.

Semiconductor Engineering
Show threadShahinDec 13, 2023
One more from #SC23 that I’m just getting to. Check out this from European Processor Initiative.
- 22nm by GF
- Long vector, RISC-V, VPU
- Variable precision, VRP, inc 128bit IEEE
- Stencil processing, RISC-V, STX, for ML
- Notably: pretty large team in multiple countries and organizations.
1.0 sampled about a year ago. Same process.
1.5 was announced just before #SC23
Andreas HertenDec 13, 2023
We are still wrapping up the year in the Lab!
In the following blog post, Chelsea writes about some tests with AI benchmarks on different hardware (A100, H100, MI250, IPU); a talk at #SC23: https://doi.org/10.34732/xdvblg-0xczeh
#HPC #GPU #AI
SC23 WHPC Workshop Paper: OpenGPT-X – Novel Architecture Exploration

The Supercomputing Conference 2023 took place in Denver, Colorado, from November 12th to 17th. For the Women in HPC workshop, we submitted a paper, which focused on benchmarking different accelerators for AI. The paper was accepted and I was invited to hold a lightning talk to show the work, spun off our OpenGPT-X project.

JSC Accelerating Devices Lab
HPC GuruDec 12, 2023

Missed #SC23?

SC has several recorded sessions on Youtube

"I am #HPC" Plenary
Opening Session
8 x Invited Talks
Awards Ceremony
Ken Kennedy, Sidney Fernbach & Test of Time Award Presentations
#Top500 BoF
#SC24 Preview

Playlist:
https://www.youtube.com/playlist?list=PLyZk_jpQ4X_rcO9Cjy2MXHMqxylftVuNh

SC23 Recorded Sessions

YouTube