You are in the main area:Events
Headerimage for: Meet Us

Meet Us

Meet Us

15. 11. 08
HLRS Presentation at SC 2008 in Austin, TX

November 15-21, 2008 High Performance Computing demonstration at SC 2008

The High Performance Computing Center Stuttgart (HLRS) is Germany's first federal HPC Center providing researchers with HPC resources. With Porsche and T-Systems we provide supercomputers to industry and have recently established the Automotive Simulation Center Stuttgart (ASCS) pooling researchers, automotive companies, hardware vendors and ISVs for precompetitive research in automotive simulation. HLRS operates a distributed computing environment based on a 72 node NEC SX-8, a 500 node IBM Blade cluster and several Linux Clusters.

At SC2008, HLRS will demonstrate its research in HPC, Visualization, and Grid Computing. We will show tools developed at HLRS (Covise, Open MPI, MARMOT) and highlight results from the TeraFlop-Workbench. Examples from industry will focus on the automotive and aerospace sectors. Scientific research is demonstrated in engineering, medicine, biology, chemistry and physics.

Scenarios and tools planned for demonstration:

Please visit us at the HLRS booth #1353. A number of projects are demonstrated that focus on designing and implementing middleware components, on putting them together to create a framework for grid computing and on running applications to exploit the potential of such frameworks.

COVISE (COllaborative VIsualization and Simulation Environment) is a visualization software that was developed in several European projects. It aims to allow distributed visualization, collaborative working and distributed simulation in a HPCN environment.

MARMOT is a MPI debugging and verification tool that aims to increase the portability of MPI programs and to track down bugs like race conditions and deadlocks.

 

Open MPI is a completely new MPI implementation based on the experience of several projects such as LAM/MPI, FT-MPI, LA-MPI and PACX-MPI, in order to build the best MPI library available, offering a complete, performant and thread-safe MPI-2 implementation.

PACX-MPI (PArallel Computer eXtension) allows to extend the functionality of native MPI implementations to clusters of MPPs and PVPs. PACX-MPI covers the complete MPI-1.2 standard.

Projects:

The Teraflop Workbench is a collaboration between NEC and HLRS. The Teraflop Workbench will enable sustained Teraflop performance for a wide range of scientific and industrial applications.

HPC-Europa is a consortium of leading HPC centers aiming at the integrated provision of advanced computational services to the European research community. The services will be delivered at a large spectrum both in terms of access to HPC systems, allowing European researchers to do science while taking advantage of HPC systems installed at their hosting center. Moreover the JRA and NA actions contribute to foster a culture of cooperation to generate critical mass for computational activities, such as seamless access to all computational ressources and advanced performance analysis of their applications.

Within the Int.EU.Grid project, a Grid is being set up, which allows Interactive access to Grid ressources, distributed in Europe. HLRS will provide support for the MPI-middleware using Open MPI and PACX-MPI.  

The BREIN project aims to enable individuals/business entities to provide their services in a controlled environment that allows intelligent and optimal management of the respective resources according to the individual business goals/constraints. For that purpose the BREIN framework will significantly reduce complexity of its usage by integrating Multiagent and Semantic Web concepts in the Grid.

The TrustCoM project is developing a framework for trust, security and contract management within dynamic virtual organisation supported by service oriented architectures implemented using Web Services and the Grid.

Demonstration Coordination at HLRS: