EU awards supercomputing strategy contract

EU wants supercomputing strategy for next 10 years

The European Commission (EC) has awarded a contract to develop a strategic agenda for high-performance computing (HPC) in Europe.

Market intelligence firm IDC will lead the consortium, comprising supercomputing experts from the French Ter@tec European HPC technology park, the UK's Daresbury Laboratory and two sites in Germany, the Leibniz-Rechenzentrum and the Forschungszentrum Jülich.

IDC's associate vice president of consulting Chris Ingle said: "This is a critical moment for HPC leadership, because Europe has been a leader in this field in the past. With the right investments, that leadership can continue to develop a strong HPC industry."

The seven-month contract will provide policymakers with a detailed analysis of HPC, focusing on HPC industry trends from 2010-20, a view on the technology requirements from the HPC industry in 2020, and a strategic agenda for European HPC.

"The link has been firmly established between HPC and scientific and economic advancement, but the investments needed for the next generation of HPC systems will be substantial," pointed out Steve Conway, IDC's technical computing group research vice president.

"Deciding on the optimal areas of investment – systems, storage, software, and people skills – that are most valuable to European HPC users, and the wider economy, is critical to the EU's success in developing its HPC agenda," added Conway.

Web site Top500 Super Computers produces monthly updates on the most powerful supercomputer sites, and November's list shows the top three sites being hosted in the US. Fourth on the list is the Forschungszentrum Jülich site, while fifth is the Chinese National SuperComputer Centre in Tianjin. The top UK site on the list is the University of Edinburgh in 20th place.

"Many countries are installing multiple petascale supercomputers today and are laying the foundation for using exascale systems in a few years," explained Conway.

Petascale systems can process one petaflop – one thousand million million floating point operations per second (flops), while exascale systems multiply that compute power by a factor of 1,000 – a million million million flops.