Budget Constrained Scheduling of Grid Workflows Using Partial Critical Paths

Workflow scheduling on utility Grids is a complex multi-objective optimization problem which tries to optimize several scheduling criteria such as execution time and cost. One common approach to the problem is to optimize only one criterion, while other criteria are constrained within fixed thresholds. In this paper, we propose a new scheduling algorithm called Budget Partial Critical Path (Budget-PCP) which aims to create a schedule that minimizes the total execution time of a workflow, such that the total execution cost does not exceed the user specified budget. The Budget-PCP algorithm has two phases: in the budget distribution phase, it recursively assigns sub-budgets to the tasks on the partial critical paths ending at previously assigned tasks, and in the planning phase it assigns the fastest service to each task while meeting its sub-budget. The simulation results show that the performance of our algorithm is very promising.