Just before the International Supercomputing Conference opens in Leipzig on Monday and at which Intel is expected to reveal more details of its HPC processors, the company has announced that two more HPC facilities have been designated as Intel Parallel Computing Centres (IPCC).
The extended IPCC programme now includes the Lawrence Berkeley National Laboratory in the USA, and the The Edinburgh Parallel Computing Centre (EPCC) in the UK.
Through a grant from Intel, the EPCC will collaborate with the company to optimise several open-source high-performance computing (HPC) applications for its latest parallel processor architectures. Similarly the Lawrence Berkeley National Laboratory will collaborate with Intel to adapt existing scientific applications to run on future supercomputers designed around many-core architectures.
The Berkeley Lab IPCC will be led by Nick Wright of the National Energy Research Scientific Computing Center (NERSC), and Bert de Jong and Hans Johansen of the Computational Research Division (CRD).
‘Although many-core processors will significantly increase supercomputing performance, that’s only part of the equation,’ said Wright, who leads NERSC’s Advanced Technologies Group. ‘To fully capitalise on this capability, we need to modernise the applications our user community uses to advance scientific discovery. Intel Parallel Computing Centres such as ours are helping to support the community to attack this problem.’
Optimising applications for many-core is important for NERSC, which announced in April that its next-generation supercomputer will be a Cray XC supercomputer using Intel’s next-generation Xeon Phi processor, which will have more than 60 cores. NERSC is working with its 5,000 users to help them adapt their codes to the new system, which will is expected to be delivered in 2016.
The Berkeley Lab IPCC will focus on increasing the parallelism of two widely used applications: NWChem and CAM5. NWChem, aims to provide its users with computational chemistry tools that are scalable both in their ability to treat large scientific computational chemistry problems efficiently, and in their use of available parallel computing resources from high-performance parallel supercomputers to conventional workstation clusters. CAM5, part of the Community Earth System Model, is widely used for studying global climat.
Modernising these codes to run on many-core architecture will enable the scientific community to pursue new frontiers in the fields of chemistry, materials and climate research. Because both NWChem and CAM5 are open source applications, any improvements made to them will be shared with the broader user community, maximising the benefits of the project.
‘Enabling NWChem to harness the full power of manycore processors allows our computational chemistry and materials community to accelerate scientific discovery, tackling more complex scientific problems and reducing the time researchers have to wait for simulations to complete,’ says de Jong, who leads CRD’s Scientific Computing Group and is a lead developer of the NWChem software. ‘Advances made by our IPCC will be shared with the developer community, including lessons learned and making our code available as open source.
The EPCC will be attempting to overcome similar challenges, increasing the parallelisation of code used for scientific applications with the EPCC’s Application Consultants undertaking the code porting and optimisation tasks. A further aim of the IPCC is to use the hardware available at EPCC, and its extensive training programmes, to provide training and expertise, on efficiently using Intel hardware for computational simulation, to a wider range of academic and industrial participants in the UK and Europe.
Professor Mark Parsons, EPCC’s Executive Director (Research and Commercialisation), said: ‘Designation as an IPCC gives us an incredible opportunity to work on a range of important, and widely used simulation codes to ensure that they can utilise the latest Intel hardware effectively. ARCHER, the UK’s national HPC service hosted and supported by EPCC is a Cray XC30 system with Intel Xeon Processor E5-2697 v2. It is therefore essential that mainstream simulation packages, which account for more than half the usage of ARCHER, are properly optimised to get maximum benefit from this technology.’
Adrian Jackson, Research Architect at EPCC and leader for EPCC’s involvement in the G8-funded Nu-Fuse project, said: ‘The centre will build on a range of world-class projects, collaborations, and initiatives that EPCC is currently involved with, including European HPC projects such as PRACE and CRESTA and global simulation initiatives, such as Nu-Fuse.’
Bob Burroughs, Director of Technical Computing Ecosystem Enabling at Intel, said: ‘Intel is pleased to expand our Intel Parallel Computing Centre program in collaborating with EPCC. This new centre creates an opportunity for Intel and EPCC to innovate and optimise applications which benefit industry and science in Europe and globally.’