Blog

From November 12 to 17, 2023, the high-performance computing (HPC) world converged at the International Conference for High-Performance Computing, Networking, Storage, and Analysis – SC23 and EXCELLERAT was no exception. With a record over 14,000 attendees and 438 exhibitors, SC23 lived up to its reputation as the premier event for scientists, engineers, researchers, educators, programmers, and developers in the HPC community.
While EXCELLERAT didn’t have a physical booth of its own, its activities were well-represented by project partners: BSC, HLRS, and SiPearl were exhibitors, and KTH was involved in the technical programme.

On Thursday, November 16, 2023, the “Day of the Slovenian Supercomputer Network” was organized by the National Competence Center SLING in Ljubljana, Slovenia. One of the biggest HPC events in Slovenia gathered visitors from the entire HPC value chain.

The varied program included presentations on the activities of the National Competence Center (NCC) and Centers of Excellence (CoEs). Furthermore, examples of the effective use of powerful supercomputers in industry and academia were presented by research institutions and companies, and the possibilities of accessing HPC infrastructure were explained.

EXCELLERAT P2 was successfully represented at the 17th International Symposium on Operations Research in Slovenia – SOR’23, held in Bled, Slovenia, September 20-22, 2023.

A simulation software often computes a configuration in a step response fashion. Indeed, it starts from an approximate initial state and iteratively advances towards a statistically converged state that satisfies the physical constraints of the model. In Computational Fluid Dynamics (CFD), for instance, the conservation laws are progressively enforced, resulting in global signals with similar characteristics. Consequently, before reaching a statistically steady state, there is a warm-up period unsatisfactory from the modeling point of view, that should be discarded in our data collection.

In this blog post, we want to bring attention to the central role of supercomputer users in the mitigation of computational waste. In a nutshell, users are not aware of wasteful behaviors. Therefore, after a brief recall of what running a supercomputer means, we will introduce new metrics for measuring computational waste, then we will describe the two main waste sources: understayers jobs and overstayer jobs. Finally we will show how we can engage users in this quest for better and cleaner workloads.

The aerospace industry can greatly benefit from using HPC and Artificial Intelligence technologies. Those technologies and significant computational power are crucial in the aerospace industry for several purposes. HPC enables complex simulations and modeling of aerodynamics, structural mechanics, and fluid dynamics. It allows aerospace engineers to perform detailed analyses of aircraft performance, including airflow patterns, stress distribution, and fuel efficiency. AI can enhance these simulations by enabling optimization algorithms and machine learning techniques to improve designs and performance.

High-performance computing (HPC) enables companies operating in any industrial sector to become more innovative, more productive and maintain a competitive edge. But above all, with the help of cutting-edge technologies such as cloud supercomputing, artificial intelligence, machine learning and big data analysis, companies can develop products and services with a higher added value. Moreover, HPC paves the way to novel industrial applications. Embracing HPC in industry to fulfil the demands for processing highly complex tasks and large volumes of data in real-time could result in significant business benefits such as reducing costs for product or service development, considerable human resources costs savings, speed up the development process and decrease time to market. Furthermore, supercomputers can process vast amounts of data in a short amount of time, allowing companies to analyse large datasets and make better-informed decisions quickly.

Have you ever tried walking in a city you’re completely new to, without having any idea where you are or how it’s organized? Would it have been easier and taken less time and effort if you had a map in the first place? Then you could have memorized the general scheme of the city, how different parts of the town are linked with each other, and you could have focused on the parts of interest to you. The idea in software geography is the same: being a new developer to a software, you could either spend months reading it linearly before figuring out how certain blocks are linked together, and finally start building a mental map of it over years – or you could start with a map.