PRACE aims at petaflop/s-class computing – CSC is part of the project
19 November 2008Tweet
One of the key machines in the developing pan-European HPC network, at the Finnish IT Center for Science, CSC, has just been updated again.
The Cray XT4/XT5 supercomputer, which has been upgraded with two Cray XT5–cabinets, forms part of the PRACE-project (Partnership for Advanced Computing in Europe). PRACE has selected broad coverage of promising architectures for petaflop/s-class systems to be deployed in 2009/2010. PRACE, funded in part by the EU’s 7th Framework Programme, is intended to create a persistent pan-European high performance computing (HPC) service and infrastructure. The service will comprise three to five petaflop/s-class HPC centres. The pan-European HPC service will be a part of the European Research Area, in which the Seventh Framework Programme is prepared to invest hundreds of millions of Euros.
CSC’s Cray XT5 -system is a joint prototype with CSCS (Swiss National Supercomputer Centre). The two Cray XT5 cabinets and CSC’s recently upgraded Cray XT4/XT5 -system have in total 10,864 computer cores and they create together a theoretical peak performance of 100 teraflop/s. This makes the new Cray XT5 system at CSC the most powerful academic supercomputer in the Nordic countries and one of the fastest supercomputers in Europe.
‘The PRACE upgrade confirms CSC’s strong position and influence in building European co-operation,’ said Kimmo Koski, managing director of CSC. ‘This powerful resource strengthens Finland's position as an attractive environment for world-class research.’
Sixteen European nations are part of PRACE. The five other prototypes will be installed at the following PRACE partner sites:
- BSC (Barcelona Supercomputing Center, Spain) has a hybrid prototype combining IBM Cell and Power6 processors. The Cell processors are used for computation and the Power6 processors for service.
- CEA (French Atomic Energy Commission, France) and FZJ (Forschungszentrum Jülich, Germany) jointly use Intel Nehalem/Xeon processors in their systems. Two shared-memory multiprocessors (thin node clusters) will be distributed over the two sites; a prototype produced by BULL at CEA and a larger system of the same architecture at FZJ.
- FZJ provides its already installed IBM BlueGene/P system, as a Massively Parallel Processing prototype.
- HLRS (High Performance Computing Center Stuttgart, Germany) offers a NEC SX-9 and an x86 based cluster as a hybrid prototype.
- NCF (Netherlands Computing Facilities Foundation, The Netherlands) evaluates the IBM Power6 architecture, a shared-memory multiprocessor (fat node cluster). This prototype will be installed in SARA Computing and Networking Services facilities in Amsterdam.