Symposium on Multi-system Application Extreme-scaling Imperative

Dirk Broemmel, Wolfgang Frings & Brian J. N. Wylie (Jülich Supercomputing Centre)



Thurdsay, 3rd September

10:00 - 11:00: 
Dirk Brömmel, Wolfgang Frings, and Brian J. N. Wylie - Extreme-scaling applications 24/7 on JUQUEEN Blue Gene/Q
Michael Bader - Earthquake Simulations at Petascale - Optimizing High Order ADER-DG in SeisSol

11:30 - 12:30
Jiaxing Qi, Harald Klimach and Sabine Roller - Performance of a lattice Boltzmann implementation on various HPC architectures
Christian Godenschwager, Martin Bauer, Florian Schornbaum, Harald Köstler and Rüde Ulrich - waLBerla, an ultra-scalable multi-physics simulation framework for piecewise regular grids

14:00 - 15:30
Martin Lanser, Oliver Rheinbach and Axel Klawonn - FE2TI: Computation Scale Bridging for Dual-Phase Steels
Aleksandr Ovcharenko, Pramod Kumbhar, Michael Lee Hines, Fabien Delalondre and Felix Schuermann - Scaling and Performance of CoreNeuron on JUQUEEN System using 1.8 Million concurrent threads

Friday, 4th September

10:00 - 11:00
Panagiotis Hadjidoukas, Diego Rossinelli, Fabian Wermelinger, Jonas Sukys, Ursula Rasthofer, Christian Conti, Babak Hejazialhosseini and Petros Koumoutsakos - High throughput simulations of two-phase flows on Blue Gene/Q
Jens Henrik Göbbert and Michael Gauding - Direct Numerical Simulation of Fluid Turbulence at Extreme Scale with psOpen

11:30 - 13:00
Nicolay Hammer, Helmut Satzger, Ferdinand Jamitzky, Momme Allalen, Alexander Block, Anupam Karmakar, Matthias Brehm, Reinhold Bader, Luigi Iapichino, Antonio Ragagnin, Vasilios Karakasis, Dieter Kranzlmüller, Arndt Bode, Herbert Huber, Martin Kühn, Rui Machado, Daniel Grünewald, Philipp V. F. Edelmann, Friedrich K. Röpke, Markus Wittmann, Thomas Zeiser, Gerhard Wellein, Gerald Mathias, Magnus Schwörer, Konstantin Lorenzen, Christoph Federrath, Ralf Klessen, Karl-Ulrich Bamberg, Hartmut Ruhl, Florian Schornbaum, Martin Bauer, Anand Nikhil, Jiaxing Qi, Harald Klimach, Hinnerk Stüben, Abhishek Deshmukh, Tobias Falkenstein, Klaus Dolag and Margarita Petkova - Extreme Scale-Out SuperMUC Phase 2, lessons learned
Miwako Tsuji - "K-scale" applications on the K computer and co-design effort for the design and development of "post-K"

In preparation for future supercomputer systems expected to employ very large numbers of processors/cores, numerous application code teams have already demonstrated the ability to exploit massive (process/thread) parallelism and adapt to limited compute-node memory on Blue Gene/Q and other HPC systems (some with Xeon Phi coprocessors or GPGPUs). This mini-symposium gives outstanding application code teams a venue to report their achievements in both strong and weak scalability, the tools they used to get insight into performance, the issues encountered which needed to be resolved, and open issues that still need to be addressed.

Submission deadline:  15th June

Notification of acceptance: 24th July

Mini Symposium: September 3-4, 2015




Last updated: 21 Aug 2015 at 10:33