The Lawrence Berkeley National Laboratory's programme to explore how cloud computing can be used to advance scientific discovery will use the System x iDataPlex server from IBM. The programme, dubbed Magellan, will be a test bed for National Energy Research Scientific Computing Division (NERSC) scientists to explore the effectiveness of cloud computing for their particular research problems. Ultimately, the project will benefit society by allowing scientists to accelerate discoveries in such disciplines as energy efficiency, climate change and genomics. The programme is funded by the American Recovery and Reinvestment Act through the US Department of Energy.
The Magellan project will use IBM's newest iDataPlex dx360 M2 server, which features double the memory and even higher power efficiency than previous versions. Harnessing iDataPlex's innovative half-depth design and liquid-cooled door, clients can lower cooling costs by as much as half and reduce floor space requirements by 30 per cent. The institute's iDataPlex system will have a theoretical peak speed of more than 60 teraflops and will be used to explore a set of possible software configurations for science clouds.
'Cloud computing has tremendous potential to accelerate scientific discoveries by making computing resources readily available to the masses of scientists. Rather than buying and managing their own cluster, a researcher can simply access a virtual cluster within the cloud. We expect many of the NERSC users to take advantage of Magellan for real scientific work, and at the same time we will be conducting research on how to build and manage science clouds,' said Kathy Yelick, head of the NERSC Division at Lawrence Berkeley National Laboratory. 'We evaluated a number of different technologies and concluded that IBM iDataPlex would deliver the best value to complete this important research.'
The Magellan research team will deploy a large cloud test bed with 5,760 processor cores on iDataPlex to evaluate a variety of scientific applications, from power grid simulations to nanoparticle analysis and analysing climate change data.
