Adaptive Grid service flow management: Framework and model

Grid computing provides the basic software infrastructure for integrating geographically distributed resources and services through standardized Grid services. One of the key challenges to enable the broader use of Grid services beyond the domain of scientific computing is the ability to perform complex tasks that require the modeling and coordination of the enactment of a number of distributed Grid services. Workflow technology is a good candidate for supporting Grid service flow. However, traditional workflow is static, thus unable to exploit the dynamic information available in the Grid and respond to the dynamic nature of the Grid. In this paper, we present an adaptive framework that provides adaptive management of Grid service flows. The framework is based on an Adaptive Grid Service Flow Model and is supported by an Event-Trigger-Rule (ETR) technology that will be used to trigger rules in a distributed fashion to adapt a Grid service flow to the dynamic Grid environment and the changing requirements of a Grid application.