The Leibniz Supercomputing Centre of the Bavarian Academy of Sciences and Humanities (Leibniz-Rechenzentrum, LRZ) is the general IT services provider of the Munich universities, one of three national supercomputing centres in Germany, and a PRACE tier-0 site. LRZ operates SuperMUC, a 6.8 Petaflop HPC system and is part of the Gauss Centre for Supercomputing (GCS) which provides the most powerful high-performance computing infrastructure in Europe.
Organization Details LRZ has been founded in 1962 as the joint computing centre for research and education for all Munich universities. We have been an active player in the area of high performance computing since more than 20 years. Besides our HPC expertise, we also provide a wide range of services to the scientific and academic communities in Munich, Bavaria, Germany, and Europe.
LRZ operates the Munich Scientific Network (Münchner Wissenschaftsnetz, MWN) that connects the Munich universities and other scientific facilities (Max Planck Institutes, Fraunhofer-Gesellschaft, etc.) in the Munich area. MWN links more than 500 buildings at over 60 sites and encompasses over 103,000 systems (servers, workstations, printers, etc.). We also provide general IT services to more than 100,000 university customers in Munich and for the Bavarian Academy of Sciences and Humanities. This includes not only internet access, but also operation of e-mail services, web servers, name servers, directory servers, and proxies. Additionally, we provide other central services, such as: compute servers, archive and storage, a Virtual Reality and Visualisation Centre (V2C), and access to special-purpose peripheral machines (e.g., CAD workplaces, large-format plotters and scanners, video editing, multimedia facilities).
HPC at LRZ LRZ offers HPC facilities on several different levels, covering the complete high performance power pyramid. Our flagship HPC system is SuperMUC, a general-purpose machine featuring only standard Intel x86 CPUs, which makes it one of the few non-hybrid HPC systems in this performance range. Thanks to its direct hot-water cooling, it is also one of the most energy-efficient supercomputers. SuperMUC delivers an aggregate peak performance of 6.8 PetaFLOP/s and provides almost 2 billion CPU hours per year to peer-reviewed research projects from Germany, Europe, and beyond.
HPC research at LRZ focuses on energy efficiency on all levels: compute nodes, system middleware, tools, applications, and building infrastructures. We are strong advocates of direct hot-water cooling, energy re-use, and holistic monitoring approaches to facilitate energy-efficient HPC operations. We contribute to multiple national and European HPC research projects, such as DEEP, Mont-Blanc, and EESI to name but a few.