IBM Blue Gene

IBM Blue Gene is a series of supercomputers developed by IBM in collaboration with Lawrence Livermore National Laboratory, University of California, Berkeley, and Columbia University. The project was funded by the United States Department of Energy and aimed to create a petaflop-scale computer to simulate complex scientific computing problems, such as climate modeling and nuclear physics, in collaboration with Los Alamos National Laboratory and Sandia National Laboratories. The development of IBM Blue Gene involved the work of renowned computer scientists, including Monty Denneau and Alan Gara, and was influenced by the TOP500 list, which ranks the world's fastest supercomputers. The project also drew on the expertise of NASA and the National Science Foundation.

Introduction

The IBM Blue Gene project was initiated in the late 1990s, with the goal of creating a highly scalable and energy-efficient supercomputer that could simulate complex scientific computing problems, such as protein folding and materials science, in collaboration with Harvard University and the Massachusetts Institute of Technology. The project involved a team of researchers from IBM Research, University of Illinois at Urbana-Champaign, and California Institute of Technology, and was influenced by the work of Seymour Cray and Cray Research. The development of IBM Blue Gene was also driven by the need for faster and more efficient supercomputers to support genomics research, such as the Human Genome Project, and astrophysics research, such as the Sloan Digital Sky Survey.

Architecture

The IBM Blue Gene architecture is based on a massively parallel processing design, using a large number of processors to achieve high performance, similar to the Cray XT3 and SGI Altix. The system uses a torus-interconnect network, developed in collaboration with Intel and Cisco Systems, to connect the processors and allow for fast data transfer, similar to the InfiniBand network used in the ASCI Purple supercomputer. The IBM Blue Gene system also features a Linux-based operating system, developed in collaboration with Red Hat and Novell, and supports a range of programming languages, including Fortran and C++, used by researchers at Stanford University and the University of Cambridge.

History

The first IBM Blue Gene system, Blue Gene/L, was developed in 2004 and was the fastest supercomputer in the world at the time, according to the TOP500 list, surpassing the NEC Earth Simulator and the ASCI White supercomputer. The system was installed at Lawrence Livermore National Laboratory and was used for a range of scientific computing applications, including climate modeling and materials science, in collaboration with University of Oxford and the European Organization for Nuclear Research. The success of IBM Blue Gene led to the development of subsequent models, including Blue Gene/P and Blue Gene/Q, which were used by researchers at University of California, Los Angeles and the University of Chicago.

Applications

The IBM Blue Gene system has been used for a range of scientific computing applications, including climate modeling, materials science, and genomics research, in collaboration with National Institutes of Health and the European Molecular Biology Laboratory. The system has also been used for astrophysics research, including simulations of black hole formation and galaxy evolution, in collaboration with NASA and the European Space Agency. Additionally, IBM Blue Gene has been used for engineering applications, such as fluid dynamics and structural analysis, in collaboration with Boeing and General Motors.

Performance

The IBM Blue Gene system has achieved high performance on a range of benchmarks, including LINPACK and HPL-AI, and has been ranked as one of the fastest supercomputers in the world, according to the TOP500 list, surpassing the Cray Jaguar and the IBM Roadrunner supercomputer. The system has also been recognized for its energy efficiency, winning the Green500 award, sponsored by Virginia Tech and the National Science Foundation. The performance of IBM Blue Gene has been influenced by the work of Gordon Bell and the ACM Gordon Bell Prize.

Models

There have been several models of the IBM Blue Gene system, including Blue Gene/L, Blue Gene/P, and Blue Gene/Q, each with improved performance and energy efficiency, developed in collaboration with AMD and NVIDIA. The latest model, Blue Gene/Q, features a powerful processor and a high-speed interconnect network, developed in collaboration with Intel and Mellanox Technologies. The IBM Blue Gene system has been used by researchers at University of Michigan and the University of Texas at Austin, and has been recognized for its contributions to scientific computing and high-performance computing, including the IEEE Computer Society and the Association for Computing Machinery. Category:Supercomputers