A scheduling algorithm for Bag-of-Tasks applications on Data Grids

In order to resolve resource scheduling problem in Data Grids to handle dynamic changes in the availability of manufacturing resource and user requirement, and to provide scalable, controllable, measurable, and easily enforceable policies for the management of manufacturing resources, we propose a Bag-of-Tasks based scheduling system. The model is then applied to present an algorithm for scheduling a Bag-of-Tasks (BOT) application on a set of geographically distributed, heterogeneous compute and data resources. Each of tasks within the application depends on multiple datasets that may be distributed anywhere within the Grid. This paper defines the problem of scheduling distributed data-intensive application on to Gird resource and presents a formal resource and application model for the problem.