IBM announced on December 6, 1999 a $100 million research initiative to build the world's fastest supercomputer, Blue Gene, to tackle fundamental problems in computational biology. The Blue Gene system will be capable of performing more than 1 petaflops per second, when built. It will achieve this performance through a combination of massive parallelism (1 million processors) and new computer architecture approaches: the system will be built through the replication of a large number of identical chips, each containing multiple processors, memory, and communication logic. Simultaneous multithreading will be used at each processor to hide
memory latency and simplify microprocessor design.

The Blue Gene project tackles fundamental problems in computer architecture and large-scale system design, such as the use of integrated processor-memory logic, multithreading, cellular design for massively parallel systems, power management, error recovery, algorithms, programming models and tools for massively parallel computing, and more. This is an ambitious, long-term research project that will push the envelope in computer science. This is also a research project that will motivate and benefit from many collaborations with academia.

Blue Gene will be used initially to study the dynamics of protein folding. Comprising strings of amino acids that are joined like links of a chain, a protein folds into a highly complex three-dimensional shape that determines its function. To understand how a protein string folds into a specific shape, we have developed a simulation model which is expected to shed significant light on this question. Such a simulation has not been feasible on current supercomputers and it will require the level of performance provided by Blue Gene In addition, the successful simulation of protein folding will require advances in the numerical modeling of molecular dynamics and in numerical algorithms. IBM Research will collaborate with the academic community in order to harness the tremendous compute power of Blue Gene to accelerate progress in computational biology and medicine.

Please contact Paridhi Verma to obtain copies of the Computer Science Brochure