The North Carolina Bioinformatics Grid will be developed in collaboration with the North Carolina Genomics and Bioinformatics Consortium (NCGBC) of more than 60 organisations representing academia, business, and biomedical research. Members include the University of North Carolina System, Duke University, GlaxoSmith Kline Inc., the Research Triangle Institute, SAS Institute, Biogen, the National Institute of Environmental Health Sciences, and many others.
"The research community will soon be overwhelmed by data resulting from the Human Genome Project and its follow-ons", stated Dr. Thomas Dunning, vice president, High-Performance Computing and Communications, MCNC. "The NC BioGrid will put North Carolina out in front of this avalanche, giving researchers and educators ready access to the genomic, proteomic, and related data being accumulated throughout the world. The BioGrid will do this so transparently that scientists will think the data resides on their workstations when, in reality, it may be distributed at many sites across the state."
Founded in 1980, MCNC is chartered to plan, develop, construct, maintain, and operate telecommunications and supercomputer facilities to support advanced education and research in North Carolina. The NC BioGrid, which strengthens MCNC's commitment to that mission, will draw heavily on capabilities of the North Carolina Supercomputing Center and North Carolina Research and Education Network (NCREN), both operated by MCNC.
IBM and MCNC will work closely with other members of the Consortium and other information technology (IT) vendors to provide the IT infrastructure, including key hardware and software technologies, and staffing to harness computing and data storage resources and make them available to NCGBC members over NCREN. These organisations will also collaborate on new Grid middleware to advance life sciences studies.
"North Carolina is at the forefront of life sciences research through the efforts of the North Carolina Genomics and Bioinformatics Consortium", said Dr. Caroline Kovac, general manager, IBM Life Sciences Solutions. "The BioGrid initiative is a model for collaborative problem solving and open computing. The new Grid will help accelerate scientific discovery by giving researchers the ability to tap computing resources on demand, ranging from applications to storage."
When completed, North Carolina's BioGrid infrastructure will include one of today's most powerful servers, the new IBM eServer p690. Based on next-generation IBM POWER4 microprocessor technology, the p690 system, code-named "Regatta", features self-healing technologies that can help provide uninterrupted operation, even through major power outages and system failures. The North Carolina system will be able to process a trillion operations per second.
IBM's Enterprise Storage Server, code-named "Shark", and Tivoli Storage Manager will provide a high-performance solution for storing, accessing, and protecting the security of research data. The Shark system configuration will support a storage area network (SAN) and is expected to provide a storage capacity of more than a petabyte of data, the content equivalent of one billion books, 400 pages each.
The BioGrid will be augmented by IBM's DiscoveryLink data integration technology which will allow researchers to easily integrate data from a variety of sources, formats and file types into a "virtual database". IBM's WebSphere e-infrastructure software will provide access and use of data via the Internet. Through the agreement, IBM and MCNC also will work together to enable a wide range of life sciences applications to run on IBM platforms and the NC BioGrid.
The first phase of the North Carolina Bioinformatics Grid Project is the construction of a test bed to test software and better understand the issues associated with storage, analysis, and movement of large bioinformatics data sets in a high-speed networked environment. The BioGrid, which will eventually scale beyond the state to other regions, is expected to be fully operational within three years.
IBM is working with the Globus open source community and others to extend Grid computing into commercial environments. In addition to working with many of the world's leading labs and research organisations in the development of Grid projects, IBM Research used Globus technologies to build its own Grid, a geographically distributed supercomputer linking IBM research and development labs in the United States, Israel, Switzerland, Japan, and England.