This level of co-operation is necessary to meet the tremendous increases in performance required for emerging genomics and proteomics applications at affordable prices, and brings together the capabilities of three experts in the fields of bio-informatics, high performance computing, and massively parallel systems. Proteomics is the study of the function, structure and interactions of proteins in cells, including humans and other organisms. "The next stage of the biotechnology revolution that was started by the Human Genome Programme will be fuelled by the successful marriage of molecular biology with high performance computing science", stated Department of Energy (DOE) Secretary Bill Richardson.
"The key aspect of this R&D relationship forms the simultaneous provision of algorithmic support, design of actual application software, and development of the system platform by three organisations with world-class competence in their respective areas", stated Bill Blake, Vice President of High Performance Technical Computing at Compaq. "This effort is a direct response to the challenge by Celera's president, J. Craig Venter, who said that even the most powerful of today's supercomputers do not meet the needs of his company's work in the genomic era. Our objective with this alliance is to apply the same full system modelling approach to bio-science that has been so successfully applied to physical sciences in DOE's National Nuclear Security Administration Stockpile Stewardship programme."
J. Craig Venter, Celera's president and chief scientific officer, commented in turn: "Just three years ago, the computational needs of biology were thought to be minor and irrelevant to the computing industry. At present, biologists are setting the pace of development for the industry. At Celera, we take pride in excelling in the application of computers to biology and the new era in medicine which is developing as a result. As Compaq Computer and the Department of Energy move toward creation of the next generation of supercomputers for defence purposes, we look forward to helping both groups develop the new machines, software and algorithms to advance life sciences."
Added Bill Camp, director of computation, computers and mathematics at Sandia, "Delivering affordable and scalable computer architectures is the foundation of modern supercomputing and has been the focus of Sandia research for more than a decade. Our knowledge will be useful because understanding the complexity of the human genome requires manipulating ever vaster amounts of information, with use of more advanced computing technologies than was required even for the assembly of the human genome itself. So we view this relationship as strategic for our continuing missions as a DOE/NNSA national security laboratory, look forward to providing world-class expertise in parallel algorithms and systems software in the cause of human health, and welcome the opportunity to play a role in developing what promises to become some of the most exciting science in recent human history."
The alliance will utilise Compaq's Alpha processors connected in massively parallel configuration with extremely high bandwidth, and low latency mesh interconnects. Compaq and Sandia will collaborate on the development of system hardware and software. Both partners have an extensive experience with supercomputers based on Alpha. Compaq already manufactures a line of supercomputers, the AlphaServer SC series, that was recently selected by the DOE's National Nuclear Security Administration as the architecture for the world's most powerful computer, the ASCI Q system, that will deliver 30 trillion operations per second when delivered in 2002. ASCI, the Accelerated Strategic Computing Initiative, is a key component of the Stockpile Stewardship programme to ensure the safety and reliability of the nation's nuclear weapons stockpile in the absence of nuclear testing.
Sandia currently operates the most powerful Linux-based supercomputer in existence, Cplant, which employs more than 1600 Alpha processors. Sandia also is home to ASCI Red, the first TeraOp supercomputer, until recently the fastest supercomputer in the world. The alliance will concentrate on future generations of the AlphaServer SC series, and the big objective is to create a prototype in the 2004 time frame. Celera and Sandia will focus on creation of advanced algorithms for biology research, as well as on new visualisation technologies for analysing the massive quantities of experimental data from high-throughput instruments. All three groups will contribute to integrating the system hardware and software and on optimising performance.
Driving the design for this next-generation supercomputer are anticipated computational and data management requirements for proteomics. These requirements are expected to be vastly more complicated than the pattern recognition and assembly operations, necessary to sequence the human genome. The researchers are counting on proteomics to take advantage of genomic databases in developing new medicines, crops, materials, and solutions to challenges in energy development and environmental cleanup.