Limelight

GCS Remains A Strong Force in HPC

The Gauss Centre for Supercomputing (GCS) resumes its status of a global heavyweight in High-Performance Computing (HPC). This is shown in the latest version of the TOP500 (7/2015) listing the fastest supercomputers in the world. Strongest GCS representative in this list remains supercomputer JUQUEEN of GCS member centre Jülich Supercomputing Centre (JSC). With a Linpack performance of 5 Petaflops (Rmax), JUQUEEN captured position 9 on the 45th edition of the TOP500, defending a rank amongst the 10 most powerful supercomputers in the world for the 4th consecutive year. Combining JUQUEEN’s compute power with the Rmax performance of the HPC systems of the other GCS member centres – SuperMUC Phase 1 and 2 (2.9 Petaflops respectively 2.8 Petaflops) of the Leibniz Supercomputing Centre (LRZ) and the 2.8 Petaflops of HPC system Hornet of the High Performance Computing Center Stuttgart – GCS now offers a total of 13.5 Petaflops Rmax to its users from science and research, which represents the by far most powerful supercomputing infrastructure in all of Europe.

“We are very proud of the fact that after all this time JUQUEEN still shines in the Top-10 of the TOP500,” comments Prof. Dr. Thomas Lippert, Director of the JSC who also serves as the GCS Chairman of the Board. “However, to us, this ranking is only one important aspect. The other is that we at GCS focus first and foremost on economical aspects and on our customers, thus system features like energy efficiency, versatile employment and sustained performance to serve our vast field of users enjoy ultimate priority,” stresses Professor Lippert. All GCS centres offer leading-edge supercomputing technologies which are designed to satisfy even the highest demands for compute performance by its users coming from all fields of science and research. Advanced cooling technologies in combination with additional sophisticated energy saving concepts ensure energy-efficient computing in all three GCS locations.

Said focus on customers and on efficiency is, for example, underpinned by the recently performed system expansion at GCS member centre LRZ, whose supercomputing infrastructure received an upgrade only in June 2015. The installation of SuperMUC Phase 2, a Lenovo NeXtScale WCT system, adds to the IBM System X iDataPlex installed at LRZ since 2012. With this move, it was decided to go with the latest technologies offering great performance while at the same time significantly cutting back on power consumption and required system footprint. SuperMUC Phase 2 delivers a Linpack performance of 2.8 Petaflops, which adds to the previously available 2.9 Petaflops of SuperMUC Phase 1. For its users, the divided LRZ system environment is transparent: SuperMUC Phase 1 and 2 are still operated like one single system as the programming environment and the file system are identical. In the current TOP500, the systems are listed separately, capturing position 20 and 21 respectively.

The GCS system infrastructure is complemented by the Cray XC40 System of the High Performance Computing Center in Stuttgart (HLRS). Code-named Hornet, the CRAY installation at HLRS will see a system upgrade later this summer, which shall result in almost doubling the now available Linpack performance of 2.8 Petaflops.

About GCS

The Gauss Centre for Supercomputing (GCS) combines the three national supercomputing centres HLRS (High Performance Computing Center Stuttgart), JSC (Jülich Supercomputing Centre), and LRZ (Leibniz Supercomputing Centre, Garching near Munich) into Germany’s Tier-0 supercomputing institution. Concertedly, the three centres provide the largest and most powerful supercomputing infrastructure in all of Europe to serve a wide range of industrial and research activities in various disciplines. They also provide top-class training and education for the national as well as the European High Performance Computing (HPC) community. GCS is the German member of PRACE (Partnership for Advance Computing in Europe), an international non-profit association consisting of 25 member countries, whose representative organizations create a pan-European supercomputing infrastructure, providing access to computing and data management resources and services for large-scale scientific and engineering applications at the highest performance level.

Click to comment

Leave a Reply

Your email address will not be published. Required fields are marked *

To Top