Improved MACO approach for grid scheduling
Parisa Rahmani, Mehdi Dadbakhsh, S. Gheisari · 2012
The grid computing system is a new trend in distributing computing systems. Grid scheduling is required in this system to achieve high performance. The goal of grid task scheduling is to achieve high system throughput and assigning the task to computing nodes. In general, there is an NP hard problem if we want to find an optimal scheduling by using the traditional sequential method. A near optimal solution can also be found by using heuristic approaches. Heuristic approaches are simpler than existing methods. The ant colony algorithm, which is one of the heuristic algorithms, suits well for the grid scheduling. In this paper we introduce a new task scheduling algorithm with load balancing based on multiple ant colony optimization (MACO). In the improved MACO approach, all colonies by using the repulsion mechanism construct their solution in parallel and find the optimal solution with a minimum execution time of task. According to experimental results, the proposed algorithm out performs the algorithms which are based on ACO.