Many complex scientific applications are modeled in the form of workflows to carry out large-scale experiments. Because of complexity of scientific processes, scientific workflows need intensive computation and data requirements. Clouds make opportunity for scientific that need high performance computing infrastructure. So scientific can run their application on cloud by their desired QoS. We propose an algorithm that able scientific to select execute plan based on their preference QoS, like time and cost. Proposed algorithm ranks the tasks in workflow and then use UPFF function for select accurate resource, based on user’s QoS. We compared our proposed algorithm with the same work by several scenarios and results show proposed algorithm has better efficiency. Keywords Scientific application, Workflow scheduling, Cloud computing
展开▼
