The Computing Services of MPCDF
The MPCDF provides facilities for high-performance computing (HPC) and data analytics/machine learning (HPDA/ML) , capacity computing (Linux clusters) and remote visualization and supports general-purpose and dedicated systems.
A list of the major computing systems can be found below. A detailed description and technical documentation can be found at the MPCDF user documentation.
The MPCDF hosts the central HPC resources of the Max Planck Society with a typical renewal cycle of 5 years. Open to scientists affiliated with Max-Planck Institutes which joined the corresponding proposals to the Max Planck Society (test access can be granted on request).
Based on Intel Xeon Cascadelake-AP processors (interim system 2020-2021): 516 compute nodes, 49,536 CPU-cores, 193 TB RAM, 3.5 PFlop/s theoretical peak (FP64), 100 Gb/s Interconnect (HDR 100, nonblocking fabric). The final system (to be deployed in two stages, by May and July 2021) is based on Intel Xeon IceLake-SP processors and Nvidia A100 GPUs.
Based on Intel Xeon Skylake-SP processors and Nvidia GPUs (V100, RTX5000): 3424 compute nodes, 136,960 CPU-cores, 128 Tesla V100-32 GPUs, 240 Quadro RTX 5000 GPUs, 529 TB CPU RAM (DDR4), 7.9 TB GPU RAM HBM2, 11.4 PFlop/s peak (FP64) + 2.64 PFlop/s peak (FP32)
Based on Intel Xeon Haswell and Broadwell CPUs and Nvidia GPUs (GTX980): 30.688 CPU cores, 128 TB RAM, 1.12 PetaFlop/s peak (FP64), 212 GPUs.
Linux Compute Clusters
The MPCDF hosts dedicated compute clusters for many Max Planck Institutes. Open to scientists affiliated with the Max-Planck Institute(s) owning the resouces.
Dedicated Linux Compute Clusters hosted for several Max Planck Institutes
Further computing infrastructure and services
Jupyter notebooks, remote visualization, remote desktops enable convenient web-based access to the HPC systems and selected linux compute clusters