Bull inaugurates its Centre for Excellence in Parallel Programming

For industrial and scientific applications to be competitive in the Petascale era of supercomputers and beyond, applications must be organised and optimised for highly parallel computer architectures. The Bull Centre for Excellence in Parallel Programming helps companies and scientists crack this massive problem: it will involve constant moving forward, towards ever higher levels of innovation and competitiveness.

Hemel Hempstead, March 22, 2013 - Bull has inaugurated its Centre for Excellence in Parallel Programming, the leading European centre of technical and industrial excellence in this field. By working with technology leaders, this centre will achieve its mission is to support engineers and scientists in research centres and industry to overcome the critical technological barrier of 'HPC application parallelisation'.

Parallel programming is all about applications' capacity to carry calculations simultaneously (hence, in parallel), without having to wait for the end of one calculation before starting another.

Parallel programming applications: an imperative for industry and science.
The next generations of supercomputers achieve performance through using many more dense configurations of processors working closely together on the same problem. The number of processors involved has moved from thousands five years ago, to tens of thousands of processors today, to millions of processors 5 years from now. 'Exascale' supercomputers at the end of this decade will likely include over a hundred million processors by taking advantage of highly parallel architectures innovations similar to today's Intel® Xeon PhiTM co-processors. According to industry analysts, only 1% of applications in use today could run 10,000 processors or more at any one time.

As a result, designers are now faced with the challenge of adapting their software if they want to benefit from the power of these innovations. Once these new supercomputers become available, only those research centres and industries that have parallelised their applications will be able to fully reap their benefits. The race has started... on a global scale.

Application parallelisation - with its smaller carbon footprint - is crucial for the development of supercomputers in the coming decades
Drawing on the expertise of a team that is unique in Europe, the Centre will be equipped with the highest levels of expertise, to help research labs and companies optimise their applications so they can be compatible with not only processors available today, but also those in development for the next generation. The Centre will supply a broad portfolio of services, including analysis and consultancy, as well as software parallelisation and optimisation.

Bull's Centre for Excellence in Parallel Programming will benefit from close cooperation with Intel in the area of supercomputers and parallel computing. It will combine the very latest Intel technologies in processing, network and storage technologies with technology from world leaders and players in research and industry specialising in parallel computing software and methodologies. Partners involved the Centre for Excellence in Parallel Programming already include Allinea, CAPS and the Joseph Fourier University.

Michel Guillemet Vice-president Bull Innovative Product commented: "The opening of this Centre for Excellence in Parallel Programming reflects Bull's determination to excel in High-Performance Computing. For many years now, we have designed and installed innovative systems to optimize supercomputers and ensure their energy efficiency; a key concern for HPC users globally. Today, with this Centre for Excellence, Bull is tackling yet another major concern for users: application parallelization, a necessity that involves significant development work."

"By moving critical applications from serial to parallel code, industry, science and other high performance computing users have the opportunity to unlock the capabilities enabling accelerated innovation" said Rajeeb Hazra, Vice-President Intel Architecture Group & General Manager, Technical Computing. "The Bull Centre for Excellence in parallel programming helps developers to fully exploit the many parallel computing features of both current and future Intel® Xeon® and Intel® Xeon PhiTM processors through standard, scalable, and portable software, providing them with compute capacity and tools needed for real innovation breakthroughs."

"On the road to Exascale Computing, many parallel programming challenges remain to be solved urgently. We need to ensure that end users from the different application domains will actually see the benefit of moving from Petascale to Exascale systems. TU Dresden is actively involved in several Exascale performance research areas. As a new Bull customer, we strongly support the Bull Centre for Excellence in Parallel Programming, and expect this initiative to both stimulate excellent research and benefit the use of computational resources at HPC client sites." commented Prof. Dr. Wolfgang E. Nagel Director of the centre for Information Services and HPC at TU Dresden.

An example:
The Pulsation project monitors the development of cyclones in the Indian Ocean, or How Parallel Programming is becoming a key environmental tool.
Certain regional natural phenomena across the planet can escape detection by scientific observers if the relevant computing models are inadequate.
Some of cold ocean currents create cloud cover that is impossible to identify if the supercomputer's computational grid is too large, with the result that simulations can deliver false results.

Parallel computing can speed up the grid calculation by a factor or ten, because it delivers ten times the power, enabling very precise simulations to be built which allow observers to identify new natural phenomena (cyclones) that could never have been detected by earlier simulations. El Niño (the cold ocean current) and La Niña (the warm ocean current) are recurring regional phenomena that were particularly difficult to predict and analyse before the High-Performance Computing applications used by weather forecasters were upgraded for parallel computing.

The particular advantage of parallelisation comes from the capacity of computing applications to process huge amounts of data simultaneously and rapidly, so that ever more accurate forecasts are possible.

Parallel programming enables applications to carry out calculations simultaneously (in parallel), with no need to wait for the completion of one calculation before starting another.

The notion of speed is directly linked to parallelisation. What use would it be, in environmental terms, if it took ten days to simulate a cyclone that is predicted to arrive in the area tomorrow?