Jülich Supercomputing Centre (JSC)

📢 FENIX 2024 Call for proposals extended until 14 February!📢 Cloud computing & archive services for your research. Europe-based researchers from any field can use FENIX services for free. Evaluation follows PRACE (Partnership for Advanced Computing in Europe) peer review. Last chance to apply:

What happens when two continents unite to shape the future of high-performance computing (#HPC)? The HANAMI High-Level Symposium held in #Barcelona showed the world the power of international collaboration. 🔎 Over 80 researchers and policymakers came together to discuss how #Europe and #Japan are teaming up to tackle the world’s biggest challenges, from #climate change to #biomedical breakthroughs. ▶️ Check out the video to see how this dynamic partnership is transforming global technology!

The successful innovation pool project on Research Software Engineering (RSE) - HiRSE_PS - will be continued until the end of 2027 as “Helmholtz Information - Research Software Engineering” (HiRSE). 🎉 HiRSE brings together all programs and centers in the Helmholtz Research Field Information to support the establishment of central RSE activities and the targeted sustainable promotion of strategically important codes by decentralized Community Software Infrastructure (CSI) groups as two mutually beneficial aspects. It also organises platforms like the HiRSE seminar series. ➡️ More information:https://buff.ly/3WtrMYJ Helmholtz Information Helmholtz-Gemeinschaft / Helmholtz Association Forschungszentrum Jülich Karlsruher Institut für Technologie (KIT)

JUNIQ Spring School on Quantum Computing The Jülich Supercomputing Centre will host the spring school on “Quantum Computing – Applications on Gate-based and Annealing Systems” from March 31 to April 04, 2025. The school focuses on hands-on experience in developing and implementing algorithms on both gate-based devices and quantum annealers. The attendees will make use of the cloud-programming platform of the Jülich UNified Infrastructure for Quantum computing (JUNIQ) to access quantum devices located at the Jülich Supercomputing Centre. The hands-on sessions will be accompanied by several lectures covering basic concepts of gate-based and annealing systems, the design and implementation of use-cases, and big-picture lectures, all given by experts in the field. The spring school will be complemented by a poster session where attendees will present their research activities to each other. Further information and the registration form can be found at: https://buff.ly/4h2u2OV Contact: training-juniq@fz-juelich.de

#JUPITER compute time exclusively awarded by the ‘Gauss AI Compute Competition’! 🎉 The Federal Ministry of Education and Research (BMBF), the Ministerium für Kultur und Wissenschaft des Landes Nordrhein-Westfalen, the Ministerium für Wissenschaft, Forschung und Kunst Baden-Württemberg and the Bavarian State Ministry of Science and the Arts have today launched a 🚩competition for computing time on the first European exascale supercomputer JUPITER. 💻 As part of the ‘Gauss-AI-Compute Competition’, which Gauss Centre for Supercomputing (GCS) is conducting together with us, researchers will have 📌 privileged access to JUPITER 📌 for a certain period of time. A total of around 15 million GPU hours ⏰ will be available for this purpose once the exascale computer goes into operation. Researchers who are successful in the competition can use these resources to train new fundamental models for #AI applications. 🤖 The Gauss AI Compute Competition is designed to support projects that effectively use JUPITER's powerful technology, while also having a high degree of societal impact. 🙌 The call for submissions ends on 👉17 March 2025. Find further information in our JSC News: https://lnkd.in/eWSeerhB #exa_jupiter #hpc #supercomputing #AI #FZJ

Join the Helmholtz GPU Hackathon 2025!   Together with NVIDIA and OpenACC, Helmholtz-Gemeinschaft (supported by Helmholtz Information & Data Science Academy) will host a GPU Hackathon from 1st-11th April 2025 at Forschungszentrum Jülich. This hybrid event is designed for researchers aiming to port or optimize their HPC or HPC+AI applications to GPUs. We will utilize our cutting-edge JUPITER hardware!   Highlights: 🤝expert mentoring for participants ⚡️exascale hardware 🚀performance gains and speedups   More info & registration: https://lnkd.in/eD-Ud3gF

Santa Clause incoming, but he forgot to bring in gift wrap. :) This is how a last minute 370 Petabyte tape delivery looks like, thanks to pro-com DATENSYSTEME GmbH and IBM. A eventful year 2024 is approaching its end, with highlights like JEDI, JETI, and the ongoing installation of #exa_JUPITER and its Modular HPC Datacenter at Jülich Supercomputing Centre (JSC)/Forschungszentrum Jülich. Grateful for having great and supportive colleagues, supporting me and our joined endeavour wherever necessary. And for having vendors bringing in more and more toys.

As part of our High-Level Symposium, we’re thrilled to host the first-ever #HANAMI High-Level Stakeholder Event focusing on: “Opportunities and Synergies for EU-Japan HPC Collaboration.” 📅 January 16, 2025 ⏰ 9 a.m. - 11:30 a.m. CET 📍 Online Here’s why you should attend: ✨ Engage with top policymakers and experts from Europe and Japan. ✨ Discover opportunities to strengthen HPC collaboration and beyond. ✨ Help shape the future of international cooperation in digital innovation. 🔗 Register now: https://lnkd.in/ga4zXvvQ ℹ️ Full agenda: https://lnkd.in/grQcHvH4 #HPC #EUJapanCollaboration #Innovation #HighPerformanceComputing #GlobalChallenges

The Heat library provides HPC infrastructure for memory-intensive applications within the NumPy/SciPy ecosystem, building on PyTorch and mpi4py. Heat v1.5 is now available!  🎉 ❓What can you use Heat for❓ ✅ port existing NumPy/SciPy code from single-CPU to multi-node clusters with minimal coding effort ✅ exploit the entire, cumulative RAM of your many nodes for memory-intensive operations and algorithms ✅ run your NumPy/SciPy code on GPUs (CUDA, ROCm, coming up: Apple MPS) Among many other new features and improvements, the main focus of Heat v1.5 is on distributed linear algebra operations, such as tall-skinny SVD, batch matrix multiplication, and triangular solver. Also introduced are vectorization via vmap across MPI processes, and batch-parallel random number generation as default for distributed operations. These new features were developed as part of the ESAPCA project, funded by the European Space Agency - ESA. ➡️ Heat on GitHub: https://lnkd.in/esQQGezN ➡️ Heat v1.5 release notes: https://lnkd.in/eC-Pn3QN ➡️ About the ESAPCA project: https://lnkd.in/eUUCzNDR Helmholtz-Gemeinschaft Helmholtz Information Helmholtz AI Deutsches Zentrum für Luft-und Raumfahrt e.V. Karlsruher Institut für Technologie (KIT) Fabian Hoppe Juan Pedro Gutiérrez Hermosillo Muriedas Björn Hagemeier Claudia Comito

🌍 Scientists in Europe: FENIX 2024 Call for proposals is open! 📢 Cloud computing & archive services for your research. Apply now & boost your project! Europe-based researchers from any field can use FENIX services. Access is free & easy. Evaluation follows PRACE peer review principles. Apply by https://lnkd.in/eNxAWFGD