Expected to be installed in February 2005, the BlueGene/L system will consist of four racks, with a peak processing speed of 22,8 trillion calculations per second, or 22,8 teraflops. The BlueGene/L supercomputer will be 24 times more powerful and use a fraction of the floor space compared to the CBRC's current computer systems. BlueGene/L consumes at most 1/10th the power per computation and 1/16th the floor space of systems of comparable compute power found on the Top 500 list of most powerful supercomputers.
"One of our biggest research challenges is to apply data obtained from genome decoding to protein engineering and drug design. The scale of simulation this requires cannot be done without the help of supercomputers", stated Dr. Yutaka Akiyama, director, Computational Biology Research Center, AIST. "IBM's BlueGene/L supercomputer provides us with a massive supercomputing resource that will dramatically accelerate our work."
The AIST scientists are developing parallel and distributed computing techniques for solving large-scale data processing and searching problems in bioinformatics. The team is also creating high performance computer applications for molecular simulation, mass spectrometry analysis, and cell simulation.
"From the beginning, IBM's commitment to developing the most powerful and flexible supercomputers in the world has included a focus on tackling the biggest challenges in life sciences", stated William Pulleyblank, director of exploratory server systems, IBM Research. "AIST's dedication to advanced life sciences research is the perfect opportunity to combine the world's most advanced supercomputer with a leading research institution to attack a major scientific challenge."
If it were installed today, this supercomputer would rank third in the world on the list of the Top 500 supercomputers announced in June 2004. Two early prototype BlueGene/L supercomputers were ranked as the fourth and eighth largest supercomputers in the world in June. These systems exploit the advanced processors based on IBM's Power Architecture. Earlier this year, IBM announced its Power Everywhere initiative, designed to make this same IBM Power Architecture more widely available for everything from consumer electronics to supercomputers.
Blue Gene is an IBM supercomputing project which is creating a new family of supercomputers optimized for bandwidth, scalability and the ability to handle large amounts of data while consuming a fraction of the power and floor space required by today's fastest systems. The project originated in 1999 in IBM's Research Division and is expected to be used worldwide by government and university researchers as well as businesses to tackle the most advanced challenges in several very different fields, including genomic research, automotive design, finance, weather forecasting and fluid dynamics. The Argonne National Laboratory in the United States and ASTRON, a leading Dutch astronomical organisation, will also be installing BlueGene/L supercomputers in 2005 to tackle unique scientific challenges.
BlueGene/L is also part of the United States National Nuclear Security Administration (NNSA)'s Advanced Simulation and Computing (ASC) Programme. IBM has partnered with the NNSA since 2001 in research and development of the BlueGene/L architecture. The NNSA will be installing a very large BlueGene/L system in 2005 at the Lawrence Livermore National Laboratory to advance understanding of materials behaviour, in particular at very high densities and temperatures.
The National Institute of Advanced Industrial Science and Technology (AIST) is an Independent Administrative Institution (IAI) under the Ministry of Economy, Trade and Industry of Japan. AIST is Japan's largest public research organisation with 25 research centres, 20 research institutes, 5 research initiatives nationwide and around 3200 employees. Aiming to become a core base for bioinformatics research in Japan, The Computational Biology Research Center (CBRC) applies advanced information science theory and large-scale high-speed computational systems to R&D efforts in order to contribute to the dramatic advance in bioinformatics and its systemization.