ICCS Members Collaborate on Award-Winning Paper

One year after Lawrence Berkeley National Laboratory and UC Berkeley established the International Center for Computational Science (ICCS) with partners at the University of Heidelberg in Germany and the National Astronomical Observatories of the Chinese Academy of Sciences in China, the first research paper submitted by ICCS-affiliated researchers will be honored with the PRACE Award. The award, sponsored by the Partnership for Advanced Computing in Europe (PRACE), will be presented to the authors at the 2011 International Supercomputing Conference.

The paper, entitled “Astrophysical Particle Simulations with Large Custom GPU Clusters on Three Continents,” will be presented on Monday, June 20, at ISC’11 in Hamburg, Germany.

“This was the very first ICCS paper, it’s pretty cool that this was selected as the 2011 PRACE Award-winning paper,” said ICCS co-founder Hemant Shukla.

ICCS was established as a bridge between China, Germany and the USA to exchange knowledge, resources and personnel to work together on new technologies, as well as enhance mutual cultural respect and understanding. The authorship of the paper reflects this international emphasis: Rainer Spurzem and Peter Berczik, Chinese Academy of Sciences and University of Heidelberg; Tsuyoshi Hamada, Nagasaki University; Keigo Nitadori, RIKEN;  and Guillermo Marcus, Andreas Kugel, Reinhard Maenner, Ingo Berentzen, Jose Fiestas, Robi Banerjee and Ralf Klessen, all of the University of Heidelberg.

According to the abstract, the authors “present direct astrophysical N-body simulations with up to six million bodies using our parallel MPI-CUDA code on large GPU clusters in Beijing, Berkeley, and Heidelberg, with different kinds of GPU hardware. The clusters are linked in the cooperation of ICCS (International Center for Computational Science). We reach about 1/3 of the peak performance for this code, in a real application scenario with hierarchically blocked timesteps and a core-halo density structure of the stellar system. The code and hardware is used to simulate dense star clusters with many binaries and galactic nuclei with supermassive black holes, in which correlations between distant particles cannot be neglected.”

Among the computing resources used by the authors was Dirac, a testbed GPU cluster at the National Energy Research Scientific Computing Center at Lawrence Berkeley National Laboratory. The cluster consists of 48 nodes (each with eight cores) with attached graphics processing units (GPUs) from NVIDIA.

ICCS began as an informal discussion three years ago, leading to the current international collaboration aimed at creating computational tools to help scientists make more effective use of new computing technologies, including multicore processors. Among the scientific fields to be engaged are physics and astrophysics, climate modeling and geosciences. ICCS brings together expertise in hardware and programming as well as scientific disciplines to address the challenge of gathering, analyzing, storing and sharing the virtual explosion of scientific data resulting from experiments and simulations.

Many current scientific applications represent years of development and were developed for computer architectures based on single-core processors. But the next generation of supercomputers is based on multicore processors and will require new programming models and algorithms. Conversely, the new processor technologies are also powering remote scientific experiments, with systems using just one multi-core processor to balance the need for computing performance with low power consumption. A primary objective of ICCS is to develop scientific solutions that are easy to adopt, energy efficient and effective in implementation across a wide range of uses.

For more information about the center, visit the ICCS website.