GENCI scoops Big Data award for 2013 from the readers of HPCwire
GENCI has just been awarded the prize for the 'Best Application of Big Data in HPC', for the use of its Curie supercomputer in the first ever simulation of the evolving structure of the entire observable Universe, carried out by the Observatoire de Paris.
The HPCwire awards are widely seen as the most prestigious recognition given by the HPC community to its own each year.
21, 2013 -
For the second year running, GENCI (the French national High-Performance Computing organization) has scooped the HPCwire Readers' Choice Award at SC'13, the annual conference dedicated to High-Performance Computing, networks, storage and analysis, in Denver (CL), USA. The winners were announced on HPCwire's stand during the course of the event and more details can be found on the HPCwire web site at www.HPCwire.com. The citation for the award presented to GENCI was as follows:
Readers' Choice Award: Best Application of Big Data, for the use of the bullx supercomputer owned by GENCI, for the simulation of the evolving structure of the entire observable Universe from the Big Bang to the present day, carried out at the Observatoire de Paris.
This unprecedented simulation - part of the DEUS (Dark Energy Universe Simulation) project aimed at gaining a better understanding of the nature of 'dark energy' and its influence on the structure of the Universe, as well as the origins of the distribution of dark matter and galaxies - was carried out on the Curie supercomputer, designed by Bull and made available to the French scientific community by GENCI (Grand Equipement National de Calcul Intensif). A world first, it involved modeling the changing structure of the WHOLE observable Universe, from the 'Big Bang' to the present day, using three cosmological models.
The simulations involved tracking the gravitational collapse of some 550 billion particles in the observable Universe, covered by over 2.5 trillion calculation points. Carrying out these simulations - which remain the most powerful cosmological simulations ever completed - required the development of an innovative and effective approach to data, to ensure that dynamic calculations were able to be executed virtually simultaneously. Ultimately, the process enabled the 150 Petabytes of raw data generated to be reduced down to just 1.5 Petabytes of refined data, which will be made available to the worldwide cosmological research community.
For more information about the DEUS project, visit www.deus-consortium.org
In particular, the results obtained will provide input data for the European Space Agency's EUCLID telescope, dedicated to observing dark matter, which is due to come into service in 2019.
The research team from the Laboratoire Univers et Théorie (LUTH, Observatoire de Paris/CNRS/Université Paris Diderot) is led by Jean-Michel Alimi1. The simulations were carried out in 2012.
The implementation of the DEUS project was made possible, in particular, by the powerful resources available at GENCI, including the Curie supercomputer which features over 92,000 processors and is capable of executing two million billion operations a second (2 Petaflop/s). The system is hosted and operated by the CEA - the French Alternative Energies and Atomic Energy Commission - at its TGCC mega Data Center in Bruyères-le-Châtel.
Catherine Rivière, CEO of GENCI, commented: "We are especially honoured and delighted to receive the Readers' Choice Award for the second successive year from HPCwire. The prize clearly illustrates the versatility of the Bull supercomputer, with its ability to execute massive simulations and to manage huge amounts of data simultaneously. It also recognizes a considerable scientific achievement, which was made possible by the close collaboration between the teams at the Observatoire de Paris, along with the assistance and expertise of the engineers at the TGCC and the support of GENCI."
1 The other members of the team are Pier-Stefano Corasaniti, Yann Rasera, Irène Balmes, Vincent Bouillot and Vincent Reverdy.
GENCI, Grand Equipement National de Calcul Intensif, is a legal entity taking the form of a société civile (civil company) under French law, owned 49% by the French State represented by the Ministry for Higher Education and Research, 20% by the CEA, 20% by the CNRS, 10% by the universities and 1% by INRIA. GENCI was created in 2007 to ensure that France achieves the highest levels in intensive computing, both at an European and international level, with the following aims:
To deliver France's policy in the area of HPC for academic research
To promote the organization of European High-Performance Computing (HPC) and participate in its actions. As such, GENCI represents France in the European PRACE project
To promote the use of computer simulation and HPC in fundamental and industrial research.
For more information, visit www.genci.fr
Press contact: Laëtitia Baudin: +33 (0)6 16 27 68 73 - firstname.lastname@example.org
Bull is the trusted partner for enterprise data. The Group, which is firmly established in the Cloud and in Big Data, integrates and manages high-performance systems and end-to-end security solutions. Bull's offerings enable its customers to process all the data at their disposal, creating new types of demand. Bull converts data into value for organisations in a completely secure manner.
Bull currently employs around 9,200 people across more than 50 countries, with over 700 staff totally focused on R&D. In 2013, Bull recorded revenues of 1.3 billion.
For more information:
http://www.bull.com / http://www.facebook.com/BullGroup / http://twitter.com/bull_com