Meeting the challenges that computing at the exascale poses requires research and development on a global scale. Worldwide a number of exascale research programmes are now being created to tackle these challenges. The European Union announced the start of its exascale research programme in October 2011 with the funding of three complementary Framework 7 research projects. The CRESTA, DEEP and Mont-Blanc projects, with a combined funding of 25 million, will each study different aspects of the exascale challenge using a co-design model spanning hardware, systemware and software applications. This funding represents the first tranche in a sustained investment in exascale research by Europe.
The three projects will be showcased at a Birds of a Feather session at SC11 in Seattle, the premier international high performance computing conference. Organised and presented by the coordinators of the three projects, they will their specific research focus and show how they complement each other and other research worldwide. The target audience for the session is the global exascale research community.
CRESTA brings together four of Europe's leading HPC centres (project lead EPCC, HLRS, CSC and PDC), a world-leading supplier of HPC systems (Cray), seven application and problem owners from science and industry (DLR, KTH, ABO, JYU, UCL, ECMWF and CRSA), Europe's leading HPC tool company (Allinea) and Europe's leading performance analysis organisation (TUD). CRESTA focuses on the use of six applications with exascale potential and uses them as co-design vehicles to develop: the development environment, algorithms and libraries, user tools, and underpinning and cross-cutting technologies. The applications have been chosen as a representative sample from across the supercomputing domain including: biomolecular systems, fusion energy, the virtual physiological human, numerical weather prediction and engineering.
The Mont-Blanc project is led by the Barcelona Supercomputing Center, and brings together leading European technology providers (ARM, Bull, and Gnodal), as well as leading Tier-0 supercomputing organisations in PRACE (Juelich, LRZ, GENCI, and CINECA). The Mont-Blanc project intends
to deploy the first generation of HPC systems built from current energy-efficient embedded technologies, and to perform the required research and development to successfully achieve exascale performance with the next generation of energy-efficient HPC systems.
Lead by Forschungszentrum Jlich, the Dynamical Exascale Entry Platform(DEEP) project proposes to develop a novel, exascale-enabling supercomputing platform along with the optimisation of a set of grand-challenge codes simulating applications highly relevant for Europe's science, industry and society. The DEEP System will realise a Cluster Booster Architecture that will serve as proof-of-concept for a next-generation 100 PFLOPS PRACE production system. In addition to the lead partner Jlich, the project partners include Intel, ParTec, LRZ, U of HD, GRS, Eurotech, BSC, MLNX, EPFL, KU Leuven, CERFACS, CYI, U of Reg CINCA and CGGVS.