Abstract:

In this paper, the scheduling of independent tasks on a tree-based grid-computing platform where resources have different speeds of computation and communication is discussed, and an integer linear programming model for the task scheduling on a fiat tree-based grid-computing platform is presented. Then, by adopting the resource fusion approach, the task scheduling on a multi-level tree-based grid-computing platform is converted into the one on a fiat tree-based grid-computing platform, and the corresponding mathematical model is proposed. Based on the proposed model, two heuristic algorithms for task allocation respectively named OPCH and OPBH are finally presented. Experimental results show that the two presented algorithms are of better performance than the First- Come-First-Served （FCFS） and the Min-min algorithms.

Key words: resource fusion, grid task, scheduling model, integer linear programming, heuristic algonthm