Volume 7, Issue 1 -
Spring/Summer 1999

Volume 6, Issue 3
Fall 1998

Volume 6, Issue 2
Spring/Summer 1998

Volume 6, Issue 1
Winter 1998

Volume 5, Issue 4
Fall 1997

Volume 5, Issue 3
Summer 1997

Volume 5, Issue 2
Spring 1997

Volume 5, Issue 1
Winter 1997

Volume 4, Issue 4
Fall 1996

Volume 4, Issue 3
Summer 1996

Volume 4, Issue 2
Spring 1996

Volume 4, Issue 1
Winter 1996

Volume 3, Issue 4
Fall 1995

Volume 3, Issue 3
Summer 1995

Volume 3, Issue 2
Spring 1995

Volume 3, Issue 1
January 1995

Volume 2, Issue 4
October 1994

Volume 2, Issue 3
July 1994

Volume 2, Issue 2
April 1994

Volume 2, Issue 1
January 1994

Volume 1, Issue 4
October 1993

Volume 1, Issue 3
July 1993

Volume 1, Issue 2
April 1993

Volume 1, Issue 1
January 1993

From the Director
Enabling Distributed Research Collaborations: HACSC, vBNS, I2 and CRPC

By Ken Kennedy, Rice University; and Lennart Johnsson, University of Houston

Bringing the University of Houston (UH) into the fold of CRPC affiliated sites fortuitously coincides with the broadening of CRPC research into high-end distributed computing (see University of Houston Becomes CRPC Affiliated Site in this issue). Having focused over the past eight years on ways to make scalable parallel computing usable, we are now turning toward a new computation model in which geographically distributed parallel computers collaborate on the solution of a new class of hybrid problems. These problems typically include multiple simulation domains, remote visualization and steering, and access to data resources around the country. They may also involve collaboration of many designers on a single project, necessitating the use of collaborative systems.

Experimentation with these computations and collaborations have been made possible by the National Science Foundation (NSF) program of connections to the experimental very high Bandwidth Networking Service (vBNS), which will eventually provide gigabit connections to all the CRPC sites, because of successful proposals submitted over the past year.

Among the first of these proposals was the one from the Houston Area Computational Science Consortium (HACSC), a consortium of three major Houston institutions (Rice University, UH, and Baylor College of Medicine). The vBNS grant HACSC won from the NSF has provided partial support for connection of the three Houston institutions to an experimental network that achieves bandwidths that are initially three times faster than typically available to universities, but later will rise to 12 to 24 times faster. While we are currently still connected at our previous speeds of 45 megabits per second, we are working together to establish a Gigapop (a value-adding, layer-three, meet point where customers can meet providers) that will eventually be the main connection for the Houston schools to the Internet 2 (I2) consortium. With more than 111 members, I2 currently includes many CRPC consortium members and affiliated sites.

The HACSC plans to concentrate on research that will increase the bandwidth capabilities of the Next Generation Internet so that full- motion video, photo-realistic data, and new applications can be transferred in real time. Large-scale experimentation on distributed computing and computationally intensive research projects will benefit the most from the HACSC vBNS connections.

UH is home to the only CAVE (a cube with four display surfaces for total virtual reality immersion) in Texas, and is the location of one of the most powerful computers in the state, an IBM SP2 with a 64-node supercomputer. UH will link virtual environments in more than one location via the vBNS, enabling collaborative research in real-time. Conceivably, teams of researchers at Argonne National Laboratory's CAVE could collaborate with researchers in the UH CAVE as they mutually manipulated an AIDS virus, seeking a cure for that disease. When more universities are equipped with Immersadesks and/or CAVEs the potential for virtual collaboration of this type will increase greatly.

Human genome sequencing and macromolecular imaging are two applications being pursued by Baylor College of Medicine and Rice University through the Keck Center for Computational Biology, which was established by a grant from the Keck foundation not long after the establishment of CRPC in 1989. In addition, researchers at Rice will use the connection to experiment with software to support geographically distributed applications and applications of multidisciplinary design optimization in the aerospace industry. Rice will also explore the computer science issues involved in networking, especially conducting quality-of-service experiments with networking and providing systems support for the high bandwidth networking.

The International Wide-Area Year (I-WAY) experiment that culminated at Supercomputing '95 in San Diego established the potential of building applications that combine the power of high-performance computing and high-bandwidth networking. With the advent of vBNS connections at all CRPC sites, we are ready to move to the next great challenge: building software and algorithms that make heterogeneous distributed computing accessible to the end user.

Table of Contents