Heuristic scheduling for bag-of-tasks applications in combination with QoS in the computational grid

The computational grid provides a promising platform for the deployment of various high-performance computing applications. One issue in implementing computational grid environments is how to effectively utilize various resources in the system, such as CPU cycle, memory, communication network, and data storage. A heuristic Qsufferage is presented to schedule the bag-of-tasks application in the grid environment. The algorithm considers the location of each task's input data, while makespan and response ratio are chosen as metrics for performance evaluation. The result of the experiment shows that Qsufferage algorithm can obtain better performance compared to the other four existing algorithms.