Dynamic service selection in workflows using performance data
Scientific Programming - Dynamic Computational Workflows: Discovery, Optimization and Scheduling
Resource Discovery Based on VIRGO P2P Distributed DNS Framework
ICCS '08 Proceedings of the 8th international conference on Computational Science, Part III
Proceedings of the 1st International Workshop on Workflow Approaches to New Data-centric Science
Hi-index | 0.00 |
In this paper we present a workflow portal for Grid applications, which supports different workflow languages and workflow optimization. We present an XSLT converter that converts from one workflow language to another and enables the interoperation between different workflow languages. We discuss strategies for choosing the optimal service from several semantically equivalent Web services in a Grid application. The dynamic selection of Web services involves discovering a set of semantically equivalent services by filtering the available services based on metadata, and selecting an optimal service based on real-time data and/or historical data recorded during prior executions. Finally, we describe the framework and implementation of the workflow portal which aggregates different components of the project using Java portlets. Copyright © 2007 John Wiley & Sons, Ltd.