A Grid Workflow Process Engine: Architecture and Simulation

Grid workflow applications are emerging as a more and more important way to manage and process large data sets, and execute scientific experiments on heterogeneous and distributed grid environment. Therefore, many efforts have been made towards the development of grid workflow management engine for grid computing. This paper proposes and simulates a new workflow process engine with hierarchical architecture. Under this architecture the workflow engine receives and processes workflows in a distributed way, but corresponds to users´ workflow submission requests and gets concrete task scheduling information centrally. Also, the engine uses a genetic algorithm (GA) based scheduler. Compared to the current engines, it´s more flexible, reliable and scalable, and also can generate more efficient solutions for workflows´ performance and minimize conflict problems that exist in grid environment.