GCS Assigns 753.6 Million Computing Core Hours to National Science Projects
BERLIN, Germany, Nov. 13 — National scientists and researchers’ demand for computing time on the high performance computing systems of the Gauss Centre for Supercomputing (GCS) continues to be unabated. The 10th GCS Call for Large- Scale Projects, which was open from July 30 to August 30, 2013 resulted in a record amount of computing time granted to ambitious German computational science and emngineering projects: The total number of 753.58 million computing core hours assigned means the largest grant of computing time ever allocated by the GCS Steering Committee. The projects awarded access to the vast GCS supercomputing resources come from a wide array of scientific fields including Astrophysics, Chemistry, High Energy Physics, and Scientific Engineering.
From the 19 applications submitted, a total of 13 national computational science projects met the strict GCS large-scale project qualification criteria and were awarded with the highly coveted computing time on the GCS high performance computing (HPC) systems. The TOP5 individual allotments of computing core hours were granted to the following outstanding projects:
Magneticum Dr. Klaus Dolag, Ludwig-Maximilians-Universität München 45M core hours on SuperMUC of Leibniz Supercomputing Centre Garching (LRZ)
Mechanochemistry of Covalent Bond Breaking from First Principles Simulations Prof. Dr. Dominik Marx, Ruhr-Universität Bochum 64.9M core hours on JUQUEEN of Jülich Supercomputing Centre (JSC)
High Energy Physics:
Lattice QCD with Wilson Quarks at Zero and Non-Zero Temperature Prof. Dr. Hartmut Wittig, Johannes Gutenberg-Universität Mainz 70M core hours on JUQUEEN of Jülich Supercomputing Centre (JSC)
2+1+1 Lattice QCD Calculations with Hex Smeared Clover Fermions Prof. Dr. Zoltan Fodor, Bergische Universität Wuppertal 65M core hours on JUQUEEN of Jülich Supercomputing Centre (JSC)
LAMTUR: Investigation of Laminar-Turbulent Transition and Flow Control in Boundary Layers – Prof. Dr.-Ing. Ulrich Rist, IAG, Universität Stuttgart 125M core hours on Hermit of High Performance Computing Center Stuttgart (HLRS)
The 13 approved large-scale projects are distributed between the three GCS HPC systems Hermit of HLRS, JUQUEEN of JSC, and SuperMUC of LRZ. All three GCS systems provide computing performance in the Petaflops-range (1 Petaflops = 1 Quadrillion floating point operations per second or: a 1 with 15 zeros) and are of complementary system design and architecture to optimally respond to the needs of the researchers, developers, and engineers. For the large-scale projects of the 10th GCS call, access to computing resources and support is granted for a time period of 12 months.
“We are very happy to see that there is a steady rise in the demand for computing time on our HPC systems,” comments Prof. Dr.-Ing. Siegfried Wagner, Chairman of the GCS Steering Committee. “GCS offers world-class HPC resources to aid in scientific computing, and this is reflected in the quality of the projects our system infrastructure is being used for. Only a couple of years ago, the now supported projects would have been impossible to accommodate as they exceeded the then available GCS resources in all aspects: the infrastructure, the software and the HPC expertise. I am proud to say that meanwhile GCS has achieved the favourable position to serve projects of this magnitude,” states Prof. Wagner who points out that, like with previous calls, GCS unfortunately could not entirely fulfil the research community’s ever increasing demand for computing power. With the 10th GCS call, almost 1.5 billion computing core hours had been requested yet only half of it– 753.6 million core hours–could be granted, primarily for lack of computing resources.
Computing time allocations for GCS Large-Scale Projects are dispersed based on scientific criteria and their technical feasibility through independent reviewers in a peer-review process led by the GCS Steering Committee. The complete list of approved GCS Large Scale Projects (10th Call) can be found at http://www.gauss-centre.eu/gauss- centre/EN/Projects/LargeScaleProjects/10th-call.html
About GCS Large Scale Projects
Per the mission of the Gauss Centre for Supercomputing, all scientists and researchers in Germany have access to the petascale HPC systems of Germany’s leading supercomputing institution. Projects are classified as “large-scale” if they require more than 35 mio. core-hours in one year on a GCS member centre’s high-end system. Computing time on the GCS systems is allocated by the GCS Steering Committee to scientifically leading, ground-breaking projects which deal with complex, demanding, and innovative simulations that would not be possible without the GCS petascale infrastructure. The projects are evaluated via a strict peer-review process on the basis of the project’s scientific and technical excellence.
The GCS Calls for Large-Scale Projects application procedure and criteria for decision is described in detail at http://www.gauss-centre.eu/gauss- centre/EN/HPCservices/HowToApply/LargeScaleProjects/largeScaleProjects_node.html
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.
Source: Gauss Centre for Supercomputing