Uncategorized

The ESPResSo summer school is a CECAM flagship school organized every year by the Institute for Computational Physics at the University of Stuttgart to train students and scientists in simulation software for soft matter physics and foster synergies between simulation experts and experimentalists. The 2026 edition focused on soft matter dynamics and porous materials (1). The event attracted 43 attendees and featured 12 talks and 17 posters. Lectures introduced the audience to particle-based simulations, polymer dynamics, active matter, Monte Carlo methods, the lattice-Boltzmann method, multiphase and reactive flows, and the dynamics of confined fluids in porous environments. In hands-on sessions, participants learned to use ESPResSo and PyStencils to simulate complex systems that resolve different time- or length-scales. The conference contributions have been collected into a book of abstracts(2). The talk slides can be obtained from the event website (1) in the “Documents” tab, and recorded lectures are available on the YouTube channel “ESPResSo Simulation Package” (3). References:

Discover an easier way to access scientific software with EESSI! Join NRIS and MultiXscale CoE for an in-person workshop at the University of Bergen introducing the European Environment for Scientific Software Installations (EESSI), instructed by MultiXscale experts Richard Topouchian and Thomas Röblitz. Pronounced “easy”, EESSI provides a common stack of scientific software installations that works across HPC systems, laptops, personal workstations, and cloud infrastructure. What will you learn?🔹 Get introduced to EESSI and its key benefits.🔹 Explore how to find and load scientific software.🔹 Gain hands-on experience running applications using the EESSI software stack. Whether you’re an HPC user, researcher, or developer, this workshop is an opportunity to discover practical ways to simplify scientific software access and improve your computing workflows. More information and registration available here

EESSI has been featured on the new podcast episode of Discovered by Compute “Netflix for Scientific Software: Inside EESSI”. MultiXscale expert and core contributor to EESSI, Kenneth Hoste from Ghent University, was interviewed at ISC Hamburg (Germany) to talk about how EESSI works under the hood. Watch the interview here:

The MultiXscale Workshop on Energy Materials and Sustainability will take place from 19 to 20 November 2026 at the National Institute of Chemistry in Ljubljana (Slovenia). The deadline for registration is open until 1 November 2026. Hosting node: CECAM-SISSA-SLING-SNS. Organisers: Energy storage materials (e.g., batteries and supercapacitors) play a pivotal role in modern energy systems by enabling the reliable use of energy and supporting the electrification of transportation. Their properties determine the performance, safety, cost, lifetime, and environmental impact. Designing and optimizing these materials requires a multiscale approach that bridges phenomena occurring across different length and time scales, from electrode materials to system-level performance. At the same time, the increasing complexity of modern energy storage systems and their underlying dynamics calls for greater computational power, making supercomputers an essential tool to perform simulations of such systems.  To address these challenges, this workshop focuses on the energy storage materials and explores how multiscale modelling and experiments can be leveraged to advance their development and optimisation. To overcome the burden of building and porting scientific software to various system architectures, including the installation of packages and the management of complex dependencies, the workshop offers an optional parallel hands-on experience on using the European Environment for Scientific Software Installations (EESSI). EESSI is a shared software stack of scientific software installations that runs on various machines and in the cloud. It supports numerous CPU and GPU targets, while providing domain scientists with a uniform experience regardless of the system they are using.  The workshop will be complemented with a guided tour of the Center for Development, Demonstration and Training for Carbon-Free Technologies. The center’s is built on cutting-edge research infrastructure in the field of battery and hydrogen technology development. Find further details through this link

MultiXscale expert Rodrigo Bartolomeu presented his work on: “Performance Portable Electrostatics for Exascale Atomistic Simulations” at the NHR Conference ’26, held in Paderborn (Germany) from 14 to 17 September 2026. He also presented a poster during the conference, contributing to the exchange of ideas around #HPC, scalable simulations, and the challenges of preparing scientific applications for exascale systems. A great opportunity to connect with the HPC community and discuss the technologies shaping the future of scientific computing! Further details available here

Rodrigo Bartolomeu, Rene Halver, Jan Meinke and Godehard Sutmann have received the PPAM Best Paper Award, in the category of PPAM 2026 Award for Workshops, for their work “Speeding up N-Body Simulation Using Low-Precision Data Types” during the 16th International Conference on Parallel Processing & Applied Mathematics (PPAM 2026), organized in Poznań (Poland) from 30 August 30 to 2 September 2026. Abstract:A large part of the performance improvements in recent Nvidia GPUs has focused on low precision floating point formats. In addition to half precision (FP16), which was already available for Volta GPUs, other floating point formats such as TensorFloat-32 (TF32) and bfloat16 (BF16) have received hardware support starting with Nvidia’s A100 card. These data types are sufficient for training neural networks and increase peak performance compared to single precision (FP32) by a factor of up to 8 or 16 for TF32 and FP16 or BF16 respectively. Most scientific applications, however, use double precision (FP64) to avoid problems due to the limited precision offered by FP32 or even lower precision formats. This prevents them from taking advantage of the increasing performance of lower precision data types. In this paper, we explore the usage of different lower precision formats for the N-Body problem to prepare for the use of tensor cores. Further information available here

The European Environment for Scientific Software Installations (EESSI) has been nominated again for the HPCwire Readers’ Choice Awards in the “Best HPC/AI Programming Tool or Technology” category. EESSI already received this award in 2024, announced during the Supercomputing conference in Atlanta (USA), and since then, It has continued to evolve significantly. In this blog post, you can find the highlights of some of the recent developments, enhancements, and milestones achieved. MultiXscale CoE had a significant positive impact on EESSI since the project started in January 2023. It has enabled to mature EESSI from a proof-of-concept to a production-ready service. The polls will close on Friday, 2 October 2026. Find further details about how to vote for EESSI here

Last Friday 25 September, MutiXscale CoE collaborated with the activity “Mini-supercomputers in your hands, made with LEGO bricks”, organized by the Computing and Advanced Technologies Foundation of Extremadura (COMPUTAEX)at Espacio UEx, Universidad de Extremadura, in Cáceres (Spain), in the framework of the European Researchers’ Night.This hands on workshop introduced young participants to the fascinating world of #HPC. They had the opportunity to build their own small supercomputers using LEGO bricks and Raspberry Pi boards, while discovering how supercomputing works and learning about the MultiXscale Centre of Excellence.A fantastic way to bring science, technology and creativity closer to the next generation of innovators! Further details available here