Keynote Speech

The 2003 International Symposium on Collaborative Technologies and Systems (CTS’03)

January 19 - 23, 2003

The Scientific Workspace of the Future: An Overview of Advanced Collaboratory and Grid Research

Rick Stevens

Director of Mathematics and Computer Science Division

Argonne National Laboratory

and

Professor of Computer Science

The University of Chicago

Abstract

In this presentation I will review the current status of Grid development in the US and discuss the future evolution of the Grid concept. Computational Grids have gained popularity as a way to organize computational and data resources via high-performance networks and advanced middleware to support new classes of scientific applications. The vision for Grids promises to empower the next generation of scientists with the computation and analysis capabilities previously only available to entire laboratories. Grid projects in the US are targeting applications areas ranging from earthquake engineering, high-energy physics, fusion, astronomy to molecular biology, ecology, oceanography and bio-engineering. Grid research and development is supported in the US by multiple federal research agencies including NASA, NSF and DOE. In 2001 NSF announced the creation of the TeraGrid, an ambitious national scale grid that will connect the three NSF supercomputing centers (NCSA, SDSC and PSC) with computing research groups at Caltech and Argonne National Laboratory to create the most capable distributing computing infrastructure for science. The resource sites included in the TeraGrid are connected by a 40 Gbps backplane network to support high-performance data transfer between sites and the ultimate virtualization of the computing resources via an integrated Grid services architecture. Grids connect not only computers and data resources, but are increasingly connecting researchers and instruments. The Access Grid project started in 1999, now has well over 100 sites, interconnect with advanced collaboration technology enable groups to work in real-time with other groups anywhere on the planet. Access Grids have been used to conduct global conferences, scientific meetings, classes, seminars and hundreds of research group meetings. Current work on the Access Grid is focusing on support for asynchronous collaboration and integration of Grid applications and software infrastructure to enable shared use of computational grids. While Grids are developing rapidly in the US, there are also significant activities in both Europe and the Asia Pacific region building and using Grids. To help coordinate the worldwide development and standardization of Grid technology, the Global Grid Forum (GGF) was established in 2000 to bring together researchers and developers of Grid software and applications to forge common Grid computing standards. The GGF will meet for the first time in Tokyo this spring. As time permits, I will also discuss the future directions of Grid research, including the various plans for deploying large-scale networks of sensors that might be used for geophysical and ecological research, and the emergence of on-demand computing Grids that could provide the computing capability for advances in personalized medicine and advanced medical research. Grids will also impact the world of entertainment, educations, and society and I will conclude my talk with some ideas for the future.

Short Bio:

Professor Stevens is the director of the Mathematics and Computer Science Division and director of Argonne's advanced computing initiative targeting the development of petaflop/s computing systems.  He is the director of The ANL/UC Computation Institute, a multidisciplinary institute aimed at connecting computing to all areas of inquiry at the University and the Laboratory. Recently he has been appointed project director for the National Science Foundation supported TeraGrid project to build the US’s most comprehensive open scientific computing infrastructure (linking ANL/UC, NCSA, SDSC and Caltech). He also heads the Futures Lab, a research group he started in 1994 to investigate problems in large-scale scientific visualization and advanced collaboration environments (his group in the Futures Lab has developed the widely deployed Access Grid collaboration system).

Professor Stevens is interested in the development of innovative tools and techniques that enable computational scientists to solve important large-scale problems effectively on advanced scientific computers. Specifically, his research focuses on three principal areas: distributed collaboration and visualization environments, high-performance computer architectures (including Grids) and computational problems in the life sciences, most recently the computational problems arising in systems biology. In addition to his research work, Prof. Stevens enjoys teaching courses on computer architecture, collaboration technology, virtual reality, parallel computing and computational science.