Navigation and service

Photo of JUWELS

Supercomputer JUWELS

In July 2018, the first module of JSC's current HPC system JUWELS (Jülich Wizard for European Leadership Science) went officially into operation at JSC. JUWELS features a versatile cluster architecture based on commodity multi-core CPUs, delivering a Linpack-performance of 6.2 petaflops (Rmax) and a theoretical peak performance of 12 petaflops. It consists of about 2,550 compute nodes with two Intel Xeon 24-core Skylake CPUs each, and about two percent of the compute nodes feature four of the latest-generation NVIDIA Volta GPUs. More: Supercomputer JUWELS …

Supercomputer SuperMUC

Supercomputer SuperMUC

The LRZ flagship computer is code named SuperMUC. With the expansion of the initial system configuration by SuperMUC Phase 2, the peak performance of LRZ’s HPC system was more than doubled to now 6.8 Petaflops. The system expansion (Phase 2) was officially taken into operation in June 2015. More: Supercomputer SuperMUC …

CRAY XC40 system Hazel Hen of HLRS

Supercomputer Hazel Hen

In October 2015, the CRAY XC40-system code-named Hazel Hen was officially taken into operation at the HLRS. Hazel Hen features 185,088 compute cores and delivers a peak performance of 7.42 Petaflop/s. Hazel Hen is used for research in the fields of scientific engineering, health, energy, environment, and mobility. More: Supercomputer Hazel Hen …