Abstract

Scientific workflows are frequently being used to model complex phenomena, to analyze instrumental data, to tie together information from distributed sources, and to pursue other scientific endeavors. Out of necessity, these complex applications need to execute in distributed environments and make use of a number of heterogeneous resources. In this paper we describe some of these applications and illustrate techniques that improve their performance and reliably in distributed environments, such as grids and clouds. Although clouds were first introduced in the business arena, they have a potential to be provide on-demand resources for scientific computations.

Date

January 1, 1970

Authors

Ewa Deelman

Journal

The International Journal of High Performance Computing Applications

Volume

24

Issue

3

Pages

284-298

Publisher

SAGE Publications