You are in the main area:Events
Headerimage for: Demonstrations at the HLRS Booth

Demonstrations at the HLRS Booth

The presentations at the HLRS booth will cover the continued collaboration with our partners at Sandia National Laboratories (SNL) and the Pittsburgh Supercomputing Center (PSC) in the framework of the G7 GIBN initiative. In the Global Wide Area Applications Testbed G-WAAT we aim to bring together computing resources and software of PSC, SNL and HLRS. An example is the coupling of the Cray T3E's at PSC and HLRS (movie) for running metacomputing applications.

Results of this collaboration will be partly visualized in a distributed virtual environment based on the work between the SNL's Interaction Laboratory and the visualization group at HLRS. Furthermore we will present European collaborations focusing on metacomputing mainly.

Scenarios planned for demonstration:

SNL will kindly provide most of the equipment for the HLRS/SNL collaboration booth. We plan to run our demos in an environment involving MPPs, workstations and a cluster of PC's. Results of some of the simulations will be visualized online. Since the demonstrations rely partly on the availability of machines at three different supercomputing centers, we will have to schedule them for a fixed date and time and will not be able to show online demonstrations on demand. These demonstrations are currently planned:

  • Advanced visualization of the Z-machine data together with SNL.
  • Fluid Dynamics Simulation of the reentry of a space vehicle (URANUS)
  • Simulations of the mechanical stability of quasi-crystals in three dimensions (IMD) :
    Abstract
  • Simulations of dilute (P3T-DSMC) and dense (P3T-MD) granular systems:
    Abstract
  • High Performance simulation of internal combustion engines (HPS-ICE).

Tools and Applications used:

The planned demonstrations will be made possible and supported by tools and applications developed at HLRS and the partner institutions:

 

PACX-MPI (PArallel Computer eXtension) aims to provide MPI programmers with a single system view of a cluster of MPPs.
 
URANUS (Upwind Relaxation Algorithm for Nonequilibrum flows of the University of Stuttgart) has been developed for the simulation of nonequilibrium flows around reentry vehicles in a wide altitude-velocity range.
 
P3T-DSMC A parallel object-oriented simulation Monte Carlo code.
 
 
IMD is a software package to perform classical molecular dynamics simulations. As such it iteratively solves Hamilton's equations of motion for a system of point masses for some time interval.
 

COVISE (COllaborative VIsualization and Simulation Environment) aims to allow distributed visualization, collaborative working and distributed simulation in a HPCN environment.
 

COVER, the COVISE Virtual Reality environment allows interactive steering and analysis of scientific data using projection based VR equipment.

Demonstration Coordination at HLRS:

  • SC98 coordination supervisor: Michael Resch
  • PACX/MPI, URANUS activities: Edgar Gabriel
  • Networking supervisor: Paul Christ
  • Networking coordination and setup: Robert Stoy
  • Visualization supervisor: Ulrich Lang
  • Visualization coordination, COVISE/COVER activities: Dirk Rantzau, Andreas Werner, Uwe Wössner.

International Collaboration Contact Points:

George Davidson/Arthurine Breckenridge
Sandia National Laboratories
Albuquerrque, New Mexico 87185-0318
505-845-8390
arbreck[at]sandia.gov

Michael Levine/Sergiu Sanielevici
Pittsburgh Supercomputing Center
4400 Fifth Avenue
Pittsburgh, PA 15217 USA
Tel. 412-268-5240
Fax: 412-268-5832
sergiu[at]psc.edu

Roland Rühle/Alfred Geiger/Ulrich Lang
HPC-Center Stuttgart (HLRS)
Allmandring 30
D-70550 Stuttgart, F.R.G.
+49-711-685-5719
ruehle[at]hlrs.de; geiger[at]hlrs.de; lang[at]hlrs.de

Public Relations/Media Relations Person:

Stuttgart/Portland

Alfred Geiger
HPC-Center Stuttgart (HLRS)
Allmandring 30
D-70550 Stuttgart, F.R.G.
+49-711-685-5719
geiger[at]hlrs.de

Stuttgart

Barbara Burr
Rechenzentrum der Universität Stuttgart (RUS)
Allmandring 30
D-70550 Stuttgart, F.R.G.
+49-711-685-5811
burr[at]rus.uni-stuttgart.de