www.gridcomputingplanet.com/news/article.php/2244781

Back to Article

IBM Linux Supercomputer To Power Life Sciences Grid
By Paul Shread
August 5, 2003

Japan's largest national research organization has ordered a powerful IBM eServer Linux supercomputer that will be integrated with other systems to form a massive, distributed computing Grid.

The IBM supercomputer will deliver more than 11 trillion calculations per second, making it the world's most powerful Linux-based supercomputer. It is expected to be more powerful than a Linux cluster currently ranked as the third most powerful supercomputer in the world, according to the TOP500 List of Supercomputers.

The supercomputer will be integrated with other non-Linux systems to form a massive, distributed computing Grid - enabling collaboration between corporations, academia and government - to support various research projects, including Grid technologies, life sciences bioinformatics and nanotechnology.

The system will include 1,058 eServer 325 systems with 2,116 AMD Opteron processors. The supercomputer will help Japan's National Institute of Advanced Industrial Science and Technology (AIST), known for its research in Grid technologies, to accelerate research using Grid technology in a wide variety of projects, including the search for new materials to be used for super conductors and fuel cell batteries and the search for new compounds that could be the basis for a cure for various malignant diseases.

The new IBM eServer 325 systems to be delivered to AIST each contain two powerful AMD Opteron processor in a 1U (1.75") rack mounted form factor. AIST will run SuSE Linux Enterprise Server 8 on the supercomputer. The Grid will incorporate the Globus Toolkit 3.0 and the Open Grid Services Infrastructure, and will link heterogeneous and geographically dispersed computing resources, including servers, storage and data, to allow researchers to collaborate through shared resources.

"The eServer 325 powered by the Opteron processor offers strong performance and extended memory addressability while ensuring backward compatibility that preserves customers' existing 32-bit software investments," said David Turek, vice president of IBM Deep Computing. "Our Deep Computing on Demand offering provides high performance computing customers the flexibility to purchase these new systems or access them on demand, only paying for the compute cycles used."

DataSynapse Grid-enables IBM Linux eServer pSeries

DataSynapse said it has Grid-enabled IBM's eServer pSeries platform for Linux with DataSynapse's GridServer.

GridServer harnesses eServer pSeries platforms to operate as part of a virtual Grid supercomputer, leveraging the scalable POWER platform to incrementally tap existing or new resources to meet application peak processing scalability requirements, DataSynapse said.

DataSynapse said the Grid enablement is being showcased as part of an IBM demonstration at LinuxWorld that has been designated a finalist for best in class solution.

"DataSynapse customers that want Linux to exploit a native 64-bit environment have the option of the most advanced 64-bit processor available today in POWER," said Chuck Bryan, IBM's director of Linux on pSeries. "These customers can now do more with less, taking advantage of Linux on POWER features such as logical partitioning to simultaneously run multiple Linux servers on a single pSeries server."

United Devices and Optive Research Expand Drug Discovery Technology

United Devices has partnered with life sciences software company Optive Research to develop software for computer-assisted drug discovery (CADD) specifically optimized for United Devices' Grid computing platform.

The partnership will allow pharmaceutical companies to speed lead identification and lead optimization phases of the drug discovery process and will result in the development of novel cheminformatic and molecular design software that was previously impossible or impractical, the companies claim.

"The most obvious benefit of supporting the United Devices Grid MP platform will be to give our customers the ability to obtain CADD results much faster than by simply using individual high-end servers or clusters," said Bryan Koontz, vice president of marketing at Optive Research. "More importantly, however, Grid computing will enable Optive to develop software that gives life science researchers the ability to do things differently and more completely, reducing the traditional compromise between speed and scientific rigor."

Optive Research was created by scientist Robert S. Pearlman to help researchers bring beneficial chemical compounds, primarily new drugs and agrochemicals, to market faster and at a lower cost. The company currently has more than 15 software products, many of them being used to reduce discovery cycle times and improve the quality of lead compounds at every major pharmaceutical company in the world, Optive said.

"Many pharmaceutical companies are starting to realize that Grid computing can be an important source of competitive advantage in drug research and development," said Piush Patel, United Devices' director of Global Grid Services. "Grid-enabling key discovery applications such as those in Optive's software portfolio is an important first step in increasing the speed and scope of molecular research."

United Devices' Grid MP Enterprise platform is now available with Optive's DiverseSolutions software for chemical diversity and library design. It will soon be available with the company's software for conformational searching and sampling (Confort) as well as de novo molecular design (EA-Inventor).