Scientific visualization is an ingredient essential to understanding the large amounts of data generated from large-scale numerical simulations, laboratory experiments and geological surveys. Visualization forms an integral component of any complete framework, together with services to handle mathematical and statistical analysis, storage, feature extraction, and other functions. To support rapid and seamless collaborations and communication between researchers across geographically disparate regions necessitates a distributed infrastructure that supports redundancy, fault tolerance, and most importantly, ease of use. We describe herein an architecture based on Naradabrokering, a publish/subscribe framework that supports the above requirements. We have implemented an initial version of this architecture and describe some initial experiments.
Bibliographical noteFunding Information:
This research has been supported by the Geophysics, Advanced Computational Research, Math-Geo, Information Technology Research (VLab project: ITR-0428774) and the Advanced Computer and Infrastructure and Research programs of the National Science Foundation, the Dept. of Energy and the Environmental Protection Agency. We would also like to thank Geoffrey Fox for useful discussions on the project.
Copyright 2011 Elsevier B.V., All rights reserved.