On-line Mapping Algorithms in Highly Heterogeneous Computational Grids: A Learning Automata Approach

Computational grid is a new paradigm in parallel and distributed computing systems for realizing a virtual supercomputer over idle resources available in a wide area network like the Internet. Computational Grids are characterized for exploiting highly heterogeneous resources; so, one of the main concerns in developing computational grids is how to effectively map tasks onto heterogeneous resources in order to gain high utilization. Two approaches for mapping the tasks exist, online mode and batch mode. In batch mode at any mapping event a batch of tasks are mapped, whereas in online mode only one task is mapped. In this paper, four on-line mode mapping algorithms based on learning automata are introduced. To show the effectiveness of the proposed algorithms, computer simulation has been conducted. The results of experiments show that the proposed algorithms outperform two best existing mapping algorithms when machine heterogeneity high.