The Bandwidth Challenge competition invites teams of technologists from the nation's most elite supercomputing facilities to push the limits of modern computer networks. The competition this year was based on the theme "serving as a model." Competitors were challenged to create methods for fully utilizing a high-speed network path to support end-to-end network applications running across a grid that included the conference's exhibit floor and the participant's home institutions using production networks. Using the IU Data Capacitor, a system designed to store and manipulate massive data sets, the IU team achieved a peak transfer rate of 18.21 Gigabits/second out of a possible maximum of 20 Gigabits/second. This performance was nearly twice the peak rate of the nearest competitor. The IU team achieved an overall sustained rate of 16.2 Gigabits/second (roughly equivalent to sending 170 CDs of data per minute) using a transatlantic network path that included the Internet2, GÉANT, and DFN research networks. "This project simultaneously pushed the limits of networking and storage technology while demonstrating a reproducible model for remote data management. Best of all, we did this using a variety of research applications that we support every day at Indiana University," said Data Capacitor and Bandwidth Challenge project leader Stephen Simms. During the competition, the IU-led team ran several cutting edge computer applications, all of which depend upon the Data Capacitor's ability to read and write data at extreme speeds. A key aspect of the demonstration was the ability to simultaneously support a mix of several different applications from the sciences and humanities, including: - Modeling and analysis of the amyloid peptide, which is thought to be the cause of Alzheimer's disease, using IU's Big Red Supercomputer, led by Mookie Baik of the IU School of Informatics and IU Bloomington Department of Chemistry.
- Live acquisition of x-ray crystallography data, led by D.F. "Rick" McMullen, of Pervasive Technology Labs at Indiana University.
- Digital preservation of ancient Sanskrit manuscripts, led by P.R. Mukund of the Rochester Institute of Technology. Performance analysis of a computational fluid dynamics application by the Technische Universitaet Dresden using its Vampir/VampirTrace software package, led by Matthias Mueller of the Center for Information Services and High Performance Computing.
- Simulations of a high energy physics reaction between the basic particles of matter, led by Scott Teige of Indiana University Information Technology Services.
"IU continues to develop deep and complementary skills in advanced networking, data storage, grids and scientific gateways," said Brad Wheeler, IU vice president for information technology and chief information officer. "This accomplishment demonstrates the cumulative expertise that we are able to apply to research problems of interest from the life sciences to humanities. "We were pleased to earn an honorable mention in the SC06 competition, and winning this year with outstanding results is a testament to the team's advanced skills," Wheeler said. The Bandwidth Challenge competition was just one aspect of IU's involvement in networking at SC07. IU was among the sponsors of the SC07 network, and IU staff helped build and manage a massive network that included more than 80 miles of fiber optic cable installed to support the SC07 conference. The Data Capacitor is powered by the open source Lustre file system and the Linux operating system. It is currently accessible to U.S. researchers though IU's participation in the TeraGrid. The Data Capacitor was developed by a team from IU University Information Technology Services, the IU School of Informatics, and Pervasive Technology Labs at Indiana University. Corporate partners for the IU Bandwidth Challenge effort include Data Direct Networks, Dell, Myricom, Inc., Force 10 Networks, Inc. The Data Capacitor project is supported in part by the National Science Foundation under NSF Award Number CNS0521433 (Craig Stewart, PI; Stephen Simms, Co-PI and project manager; Caty Pilachowski, Randall Bramley and Beth Plale, Co-PIs). IU's involvement in the TeraGrid is supported in part by NSF grants ACI-0338618l, OCI-0451237, OCI-0535258, and OCI-0504075. IU's Big Red Supercomputer was funded in part by a grant from the Lilly Endowment, Inc. for the Indiana METACyt Initiative. Any opinions, findings, and conclusions or recommendations expressed in this material are those of the authors and do not necessarily reflect the views of the National Science Foundation or the Lilly Endowment Inc.