SC2002 Bandwidth Challenge Winners Set New Network Performance Marks

November 27, 2002

Prizes for most daring networking awards sponsored by Qwest Communications

November 27, 2002--The High-Performance Bandwidth Challenge awards announced last week at SC2002 established new records for network performance, with one winning application demonstrating a five-fold increase over the previous top-mark set last year. The High-Performance Bandwidth Challenge, held at the annual SC Conference, is a competition for leading-edge network applications developed by teams of researchers from around the world.

For the third consecutive year, a team led by Lawrence Berkeley National Laboratory captured the competition for the "Highest Performing Application" with a wide area distributed simulation using Cactus, Globus and Visapult software demonstrating a peak data transfer rate of of 16.8 gigabits per second, nearly 25,000 times faster than a typical home broadband connection. The winning team included the Albert Einstein Institute, Argonne National Laboratory, ESnet, Force10 Networks,Masaryk University, the National Center for Supercomputing Applications (NCSA), the National Energy Research Scientific Computing Center (NERSC), Poznan Supercomputing and Networking Centers, Sandia National Laboratories, and the University of Amsterdam.

The Data Reservoir application, demonstrated by Fujitsu Laboratories and University of Tokyo, won the award for the "Most Efficient Use of Available Bandwidth," with a peak of 585 megabits per second.

With a peak transfer rate of 2.4 gigabits per second, the award for the "Best Use of Emerging Network Infrastructure" went to Project DataSpace, which was demonstrated by a team from CANARIE, the National Center for Data Mining at the University of Illinois at Chicago, Northwestern University, SARA (Stichting Academisch Rekencentrum) and StarLight.

Ten outstanding entries third edition of the High-Performance Bandwidth Challenge was again sponsored by Qwest Communications and attracted Additional network monitoring and measurement support was provided by CalNGI, Internet2, the National Partnership for Advanced Computational Infrastructure, the San Diego Supercomputer Center, Spirent Communications, and the University of Florida.

The competition took place live on the SC2002 show floor in Baltimore using SCinet, the state-of-art, on-site network designed and built especially for the annual SC conference. SCinet featured more than 40 gigabits per second (Gbps) of external network capacity, itself a new record and more than 50,000 times the speed of a typical home broadband connection. For more information about this year's High-Performance Bandwidth Challenge, see

About SC Conferences
The annual SC Conferences bring together scientists, engineers, visualization artists, programmers, and managers to share ideas and to glimpse the future of high performance networking and computing, data analysis and management, visualization, and computational modeling. The conference is sponsored by the Institute of Electrical and Electronics Engineers Computer Society and by the Association for Computing Machinery's Special Interest Group on Computer Architecture. SC2003 will be held in Phoenix, Arizona For more, see

Last Updated: 17 February 2010