On the load distribution and performance of meta-computing systems

In this paper, we study a high-performance Heterogeneous Distributed System (HDS) that is employed as a computing platform or grid. Precisely, we study the problem of scheduling a large number of CPU-intensive tasks on such systems. In this study, the time spent by a task in the system is considered as the main issue that needs to be minimized. The proposed techniques of scheduling dynamic tasks consist of two heuristic algorithms; Recursive Neighbor Search (RNS) and Augmented Tabu-Search (ATS) algorithm. Our technique does not address directly the load-balancing problem since it is completely unrealistic in such large environments, but we will show that even a nonperfectly load-balanced system can behave reasonably well by taking into account the tasks' time demands. These algorithms are compared to a well known scheduling algorithm, in order to compare, evaluate, and clarify their performance.