Dr. Ing. Bernd Mohr, a senior scientist at the Juelich Supercomputing Centre (JSC), also serves as an ISC Executive Consultant. The International Supercomputing Conference (ISC) developed from a small expert workshop in Mannheim back in 1986 into a full-blown five-day technical HPC conference today. Now in its 29th year, ISC is the world’s oldest and Europe’s most important conference and networking event for the HPC community. What bugs me is I still run into people who make their living in the HPC or the broader scientific computing world, but think it is not worth going to the conference because they went there a decade ago and got the impression it is just a vendor trade show with some big-name keynote talks. Or my favorite: some friend of theirs attended a decade ago and told them about it … but they were never there themselves.

If that describes you, then you are a little bit like these guys who had a look at Fortran back in 1990s and since then tell everybody it is an old, arcane programming language useless in today’s modern world — thereby ignoring all the important development and enhancements introduced by Fortran 1990, 1995, 2003, 2008, … I think you get it. Like many things in this world, me included ;-), the ISC conference did not just get older over the years; no, it became more mature, experienced and sophisticated!

A clear example of this is the very strong research program at ISC. It developed from a small research workshop that used to take place a day before the actual conference. At that time, it consisted of a few selected papers that were published in a journal years after the event.

Today, this is a full-blown, single-track research program completely integrated into the main conference. Getting a paper accepted for presentation and into the ISC proceedings, which now get published by Springer in conjunction with the conference, is increasingly competitive. Of course, besides the research papers, there are also expert peer-reviewed tutorials, posters and — coming in 2015 — workshops.

An interesting facet of the research program is the awards for selected papers. Originally, there was a conference award supported by the main event sponsor of that year. However, since 2008, the European HPC community, in the form of PRACE, has awarded the best paper representing a breakthrough in science achieved through high performance computing. The paper must describe an algorithm or implementation that achieved a significant improvement in scalability or performance, or a novel approach to performance evaluation on a massively parallel architecture. The winner is given the opportunity to deliver a presentation about their work during the Monday afternoon research papers session.

Past PRACE awards winners include

  • 2008: “UCHPC – UnConventional High Performance Computing for Finite Element Simulations,” Stefan Turek, Dominik Göddeke, Christian Becker, Sven H.M. Buijssen and Hilmar Wobker; Institute of Applied Mathematics, Dortmund University of Technology, Germany
  • 2009: “High Scalability Multipole Method. Solving Half Billion of Unknowns,” J. C. Mouriño, A. Gómez, J. M. Taboada, L. Landesa, J. M. Bértolo, F. Obelleiro and J. L. Rodríguez; Supercomputing Center of Galicia (CESGA), Universidad de Extremadura, Universidad de Vigo, Spain
  • 2010: “Massively Parallel Granular Flow Simulations with Non-Spherical Particles,” Klaus Iglberger and Ulrich Rüde; University of Erlangen-Nuremberg, Germany
  • 2011: “Astrophysical Particle Simulations with Large Custom GPU Clusters on Three Continents,” R. Spurzem, P. Berczik, T. Hamada, K. Nitadori, G. Marcus, A. Kugel, R. Manner, I. Berentzen, J. Fiestas, R. Banerjee and R. Klessen; Chinese Academy of Sciences, Nagasaki University, RIKEN, University of Heidelberg
  • 2012: “A Fast & Scalable Low Dimensional Solver for Charged Particle Dynamics in Large Particle Accelerators” Yves Ineichen, Andreas Adelmann, Costas Bekas, Alessandro Curioni, Peter Arbenz; Paul Scherrer Institut, IBM Research – Zurich and ETH Zürich, Switzerland
  • 2013: “591 TFLOPS Multi-Trillion Particles Simulation on SuperMUC,” Wolfgang Eckhardt, Alexander Heinecke, Reinhold Bader, Matthias Brehm, Nicolay Hammer, Herbert Huber, Hans-Georg Kleinhenz, Jadran Vrabec, Hans Hasse, Martin Horsch, Martin Bernreuther, Colin W. Glass, Christoph Niethammer, Arndt Bode, Hans-Joachim Bungartz; Technische Universität München, Leibniz-Rechenzentrum der Bayerischen Akademie der Wissenschaften, University of Paderborn,  Laboratory of Engineering Thermodynamics (LTD), TU Kaiserslautern, and High Performance Computing Centre Stuttgart (HLRS), Germany

In addition, since 2011, the German HPC community, represented by the Gauss Center for Supercomputing (a coalition of the German national supercomputing centers at Garching, Jülich and Stuttgart), has presented an award for the most outstanding paper in the field of scalable supercomputing. The 2014 winner will be awarded 3,000 Euros and be given the opportunity to deliver a presentation during the Monday afternoon research papers session.

Past Gauss winners include

  • 2011: “Experiments with the Fresh Breeze Tree-Based Memory Model,” Jack Dennis, Guangrong Gao,  Xiaoxuan Meng; Massachusetts Institute of Technology and University of Delaware
  • 2012: “Blue Gene/Q: By Co-Design,” The Blue Gene Team; IBM Systems and Technology Group, Rochester, MN, USA
  • 2013: “TUE, a New Energy-Efficiency Metric Applied at ORNL's Jaguar,” Michael K. Patterson, Stephen W. Poole, Chung-Hsing Hsu, Don Maxwell, William Tschudi, Henry Coles, David J. Martinez, Natalie Bates; Intel Architecture Group, Intel Corporation,  Oak Ridge National Laboratory, Lawrence Berkeley National Laboratory, Sandia National Laboratories, and Energy Efficient HPC Working Group, USA

Find out who won the PRACE and Gauss Awards this year. The awarding takes place on Monday, June 23, right after the ISC’14 conference opening session.

Dr. Ing. Bernd Mohr, a senior scientist at the Juelich Supercomputing Centre (JSC), also serves as an ISC Executive Consultant. He may be reached at