Published 05/15/2000
Contact: David Hart, SDSC, dhart@sdsc.edu, 858-534-8314.
SAN DIEGO, CA - The San Diego Supercomputer Center (SDSC) has released version 1.1.7 of the SDSC Storage Resource Broker (SRB) software package, which provides a uniform interface for managing data from distributed, heterogeneous storage resources. Version 1.1.7 supports integration with the Grid Security Infrastructure (GSI) for authentication and DataCutter software for data extraction.
The GSI authentication scheme uses an X.509 public key infrastructure like those used in Web-based e-commerce to securely identify users and servers across the Internet and across administrative domains. The DataCutter software from the University of Maryland extracts desired subsets of data at the location where the data is stored, returning only the requested information -- an important savings in network traffic when large data sets are involved.
Interest is growing in the innovative SDSC SRB software, developed by SDSC's Data-Intensive Computing Environments (DICE) group, because of the need to integrate, manage, and access the explosively growing data collections in astronomy, physics, biology, environmental science, museums, medical records, and many other fields. There are currently more than 140 registered users of the SDSC SRB.
"The SDSC SRB software resides on the user's system and works in conjunction with a Metadata Catalog (MCAT)," says researcher Arcot Rajasekar of the DICE group. "Together the SDSC SRB and MCAT find the desired information based on a set of descriptive attributes, both system-level and application-specific, rather than requiring the user to know all the file names and machine locations."
An example of the SDSC SRB's capabilities is that sustained data transfer rates of 64 megabits per second have been achieved between SDSC and Caltech. Further SDSC SRB refinements in version 1.1.7 include an extended container system for supporting multiple cache and archival storage resources. Additional metadata support now handles such features as annotations and user-controlled metadata.
The DICE group has also used the SDSC SRB in projects as diverse as developing persistent archive prototypes for the National Archives, helping a group of museums in building a distributed online catalog of images of their collections, and integrating diverse types of environmental data.
"In the integration of environmental data, for example," Rajasekar says, "each user contributes their data while maintaining control over their own data sets, but sharing it with other users as needed, resulting in rapid new collaborations that simply couldn't be done before."
The GSI authentication in SDSC SRB version 1.1.7 is the third supported authentication scheme. GSI was developed as part of the Globus project, led by Carl Kesselman at the University of Southern California Information Sciences Institute and Ian Foster of Argonne National Laboratory. The Globus team participates in NPACI's Metasystems thrust area.
The DataCutter middleware suite for subsetting, indexing, and filtering data was developed by NPACI participants Joel Saltz, Alan Sussman, and Tahsin Kurc at the University of Maryland.
DICE researchers on the SDSC SRB project include Rajasekar, Michael Wan, Bing Zhu, and George Kremenek. For more information on the SDSC SRB see http://www.npaci.edu/DICE/srb.
The San Diego Supercomputer Center is an organized research unit of the University of California, San Diego, and the leading-edge site of the National Partnership for Advanced Computational Infrastructure ( http://www.npaci.edu/). SDSC is funded by the National Science Foundation through NPACI and other federal agencies, the State and University of California, and private organizations. For additional information about SDSC and NPACI, see http://www.sdsc.edu/ or contact David Hart, dhart@sdsc.edu, 858-534-8314.