CPU-based high-performance computing system

• Subject Area: Computer Science
• Term: since 2026
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 585113217

The high-performance computer (HPC) to be acquired is a standard system, similar to those found at other universities of comparable size and with similar research areas. It is intended for exceptionally complex computing tasks carried out by researchers at all institutes. The average number of users is around 40-50 active users per month, and the trend is rising. The researchers mostly come from the faculties of electrical engineering, natural sciences, mechanical engineering, mathematics, and process engineering. Access to the HPC is provided via the OVGU network infrastructure, so that the system is accessible and usable by the entire campus and thus by all scientists. Access from outside the internal network is provided via VPN. There are three research groups that should be highlighted among the users from all research areas at OVGU. The Chair of Mechanical Process Engineering at the Institute of Process Engineering works on understanding the physical behavior of particles in flows and developing computer models to predict the interactions between particles and flows. The Chair of Fluid Mechanics and Fluid Engineering at the Institute of Fluid Engineering and Thermodynamics focuses on research into two-phase flows, medical flows, reacting flows, computational fluid dynamics (CFD) and optimization, fluid machines, and non-Newtonian flows. The Chair of Heat- and Mass-Transfer at the Institute of Fluid Mechanics and Thermodynamics primarily works with molecular dynamic simulations. Matter is modeled on an atomic scale based on molecular interactions. Challenges include the need to simulate a large number of particles over a long period of time, which requires significant computing resources. All three working groups require high to very high CPU-based computing power for their simulations. Furthermore, there are many working groups and individual researchers who rely less frequently on the computing power of HPCs and also make use of them. The proposed device configuration is specifically designed to meet these requirements. Fundamentally, all of these work groups need CPU-based computing power. Therefore, a computing cluster with powerful compute nodes and a fast connection network is to be procured. The system is expected to achieve with 72 compute nodes a peak computing power of 943 TFlop. The scale is based on the good utilization of the current system, so that the current computing capacities can be guaranteed in the future.

• DFG Programme: Major Research Instrumentation
• Major Instrumentation: CPU-basierter Hochleistungsrechner
• Instrumentation Group: 7040 Vektorrechner
• Applicant Institution: Otto-von-Guericke-Universität Magdeburg
• Leader: Dr.-Ing. Stephan Jacob