GCS Awards 703 Million Computing Core Hours for Large-Scale Simulations
Newsflash 11/2019 –

With the 22nd GCS Large-Scale Call, the Gauss Centre for Supercomputing (GCS) scientific steering committee approved the allocation of 703 million core hours of computing time to eleven scientifically outstanding German research projects relying on the support of petascale-performance high-performance computing (HPC) technology.

The projects, which will run on HPC systems installed at the High-Performance Computing Center Stuttgart (HLRS), the Jülich Supercomputing Centre (JSC), and the Leibniz Supercomputing Centre Garching (LRZ), come from the fields of astrophysics and astronomy, biomedical technology and medical physics, elementary particle physics, experimental condensed matter physics and mechanics and fluid mechanics.

All of the approved projects taking part in the 22nd GCS Large-Scale Call fulfilled the strong qualification criteria set out by the GCS scientific steering committee. The three GCS centres support complementary architectures, empowering the committee to match project proposals with the system that best suits researchers’ needs. In this call, however, one research project in particular will benefit from having access to multiple GCS HPC systems, giving the team the ability to tailor their workflow to different machines’ technical specifications. The project “The QCD Phase Diagram in the Quark Mass Plane” under leadership of Prof. Dr. Zoltan Fodor (Bergische Universität Wuppertal) will be supported with 65 million core hours of computing time on the LRZ supercomputer SuperMUC-NG, an Intel Xeon processor-based HPC system, and with additional 15 million core hours on JUWELS, the multi-core processor based HPC cluster hosted at JSC.

Computing time for all simulation projects approved with the 22nd GCS large-scale call started on November 1st and will last until October 30th, 2020. The three projects granted computing time at HLRS—among them a project led by Dr. Matthias Meinke of the Institute of Aerodynamics at RWTH Aachen University, which was awarded with the largest individual computing time allocation of this call (215 million core hours)—will start out with using HLRS’s current supercomputer Hazel Hen. Currently, the Cray XC40 system is in its final phase of operation, and HLRS staff is beginning the transition to the centre’s next-generation supercomputer Hawk, an AMD processor-powered HPC system provided by Hewlett-Packard Enterprise (HPE). Once Hawk has gone into operation at HLRS—which is scheduled for the middle of next year—-the remaining computing time allocations of these three projects will be transferred from Hazel Hen to Hawk.

The complete list of the 22nd GCS Large-Scale Call projects can be found here.

November 6, 2019

Tags: Large-Scale Project Award HLRS LRZ