Efficient Cost Optimization for Workflow Scheduling on Grids

Complex applications in e-science and e-business are usually modeled as workflows, which often require the power of the Grid to run efficiently. Cost optimization for workflow applications with deadline constraints is a fundamental problem on Grids. In this paper, a workflow scheduling algorithm that minimizes execution cost while meeting the deadline is proposed. The proposed solution consists of two steps: first, an integer programming (IP) approach is applied to find the optimal decomposition of global deadline constraints into time windows for all tasks. Second, we use local selection to find the best resources that satisfy these local time window constraints. The results of experimental evaluation show that our approach significantly outperforms existing solutions in terms of workflow execution cost while meeting the deadline constraint.