Abstract
The coordinated use of geographically distributed computers, or
metacomputing, can in principle provide more accessible and
cost-effective supercomputing than do conventional high-performance
systems. However, we lack evidence that metacomputing systems can be
made easily usable or that large numbers of applications
are able to exploit metacomputing resources. In this article, we present
work that addresses both these concerns. The basis for this work is a
system called Nimrod that provides a desktop problem-solving
environment for parametric experiments. We describe how Nimrod has
been extended to support the scheduling of computational resources
located in a wide-area environment and report on an experiment in
which Nimrod was used to schedule a large parametric study across the
Australian Internet. The experiment provided both new scientific
results and insights into Nimrod capabilities. We relate the results
of this experiment to lessons learned from the I-WAY distributed
computing experiment and draw conclusions as to how Nimrod and
I-WAY--like computing environments should be developed to support
desktop metacomputing.