Modeling, Optimization and Performance Evaluation of Scientific Workflows in Clouds

Although clouds offer new opportunities for scientists to access compute resources on demand, the existence of a large number of infrastructures with different billing schemes precludes well-established resource allocation policies, known e.g. From grids. Our research aims to provide methods and tools for cost optimization of cloud-based scientific workflows under various constraints, imposed either by the user (e.g. Deadlines) or by the infrastructure (e.g. Limited available instances).