Équilibrage de charge dynamique dans les grilles de calcul : une approche à base d'agents

Abstract

In grid computing, if the system load in each of resources is nearly equal, it indicates correct use of resources. In this thesis , an agent based load balancing model is presented. In addition of balancing the load in grid by job migration technique , or by moving an entire process to a underloaded resources. The proposed agent based load balancing model aims to take benefit of the agent‟s characteristics to generate an autonomous system. It also has a distinct advantage over other agent based load balancing models on resource utilisation. The performance evaluation appears that the proposed algorithm can enhance the overall performance of grid computing