June 22–26, 2014
Leipzig, Germany

Session Details

Name: Advanced Re-Engineering of HPC Applications
Time: Wednesday, June 25, 2014
11:30 am - 01:00 pm
Room:   Hall 5
CCL - Congress Center Leipzig
Breaks:01:00 pm - 02:15 pm Lunch
Chair:   Achim Basermann, DLR
Abstract:   Today´s HPC systems combine massive parallelism on system level with massive parallelism within many-core processors and special accelerators like FPGAs or GPGPUs. On the software side, this requires deep re-engineering of HPC application codes and also deep changes of the basic mathematical algorithms applied. Trends in massively parallel HPC applications are, among others, the use of particle methods, application of PGAS programming models as well as employing elaborate data partitioning and communication reducing methods. The presentations in this session show how application codes from weather forecasting, haemodynamics and rotor wake simulation meet performance and scalability challenges on massively parallel hardware.
Presentations: Challenges of Getting ECMWF's Weather Forecast Model (IFS) to the Exascale
11:30 am - 12:00 pm
  George Mozdzynski, ECMWF
Optimizing Haemodynamics Simulation for Supercomputers: Coalesced Communication & Weighted Decomposition
12:00 pm - 12:30 pm
  Derek Groen, UCL
Rotor Wake Simulation via Vortex Lattice Methods on a Workstation Using GPGPU Accelerators
12:30 pm - 01:00 pm
  Johannes Hofmann, DLR