ACADEMIA

SGI's Storage Solution Enables PSC to Store and Manage more than a Petabyte of Data - Scientific Director, Mike Levine, to Speak on Terascale Project at First SGI Users Conference -- SGI today announced that the Pittsburgh Supercomputing Center (PSC) has selected an SGI® storage solution to help the center meet its growing storage challenges, with plans to ultimately store and manage over a petabyte of data. A petabyte is about one-hundred times the contents of the Library of Congress, the largest library in the world, with more than 18 million books, 2.5 million recordings, 12 million photographs, 4.5 million maps, and 54 million manuscripts. Pittsburgh Supercomputing will use SGI® Data Migration Facility(tm) (DMF), the company's hierarchical storage management software, to manage all data generated by its Terascale Computing System, currently used by over 1,000 researchers worldwide. PSC is also replacing an existing CRAY® J90 system and its data repository, and will migrate the data to a new SGI® Origin(tm) 300 system running DMF. The SGI system will be integrated into the terascale computing environment, augmenting the most powerful facility in the United States committed solely to public research areas including earthquake preparedness, AIDS research, storm prediction and protein folding for biotechnology and pharmaceutical applications and is open to all science and engineering disciplines. Cost Benefits and Simplified Data Transition PSC chose to migrate to an SGI® storage solution for a number of reasons. First, the data will be migrated to the DMF environment without changing the data format of the existing environment, requiring only a database conversion and no change to the data itself. This seamless transition is expected to result in significant savings in time and money, both critical issues when managing vast amounts of data. In addition, the Origin(tm) system is compatible with the UNICOS® OS-based storage tapes that were used by the previous system. The new Origin system will interface with the existing tape libraries, while DMF will manage the data placement and migration policies to fit the researchers' requirements. Another benefit of the SGI solution is the ability to utilize existing tape media containing data originally formatted by the Cray system. Previously recorded data saved on tape media can be read, restored, and managed without conversion, eliminating the time-consuming process of recopying all the data in a new format. SGI's ability to deliver a secure and seamless transition of data to the new system was clearly important to PSC. "SGI has proven expertise in delivering data management solutions for the massive amounts of data typically generated in the work environments of technical, scientific, and creative communities. DMF gives us the high-performance data flow and storage management we need, with a high level of security and scalability. Together with SGI, the terascale project and PSC will better enable our over 1,000 researchers to conduct leading-edge scientific research," said Mike Levine, scientific director, Pittsburgh Supercomputing Center. "Through our relationship with the Pittsburgh Supercomputing Center, SGI will bring the benefits of virtually unlimited storage capacity to researchers worldwide," said Bob Bishop, chairman and chief executive officer, SGI. "The increasing complexity of cutting-edge research data makes our relationship with PSC imperative." Levine will speak on the terascale project at the first 2003 SGI Technical Users' Conference, held June 11-13, 2003, at SGI corporate headquarters in Mountain View, Calif. The conference will bring together SGI customers who are on the forefront of high-performance computing and collaborative visualization-visionaries who rely on SGI® solutions to meet their most mission-critical scientific, engineering and creative challenges. Investment Protection By leveraging the capabilities of SGI systems and the expertise of SGI's professional services group, PSC will integrate the new storage hardware into this system with relative ease and without inconvenience to the user community, protecting the center's initial investment in this system architecture. The TeraGrid Project The upgraded system storage will also help the PSC in its role as a member of the TeraGrid project. The TeraGrid was "lit-up" for the first time at the end of February and is the world's largest, fastest, distributed infrastructure for open scientific research. The TeraGrid includes 20 teraflops of computing power distributed at five sites and will also include high-resolution visualization environments, and toolkits for grid computing. These components will be tightly integrated and connected through a network that will operate at 40 gigabits per second - the fastest research grid on the planet. Other participants in this revolutionary grid computing project include the National Center for Supercomputing Applications at the University of Illinois, Urbana-Champaign, the San Diego Supercomputer Center at the University of California San Diego, Argonne National Laboratory in Argonne IL, and the Center for Advanced Computing Research at the California Institute of Technology in Pasadena.