Research of Task Scheduling of Ad hoc Cluster Computing Based on Genetic Algorithm

The system of ad hoc cluster computing is a flexible, self-organization and expandable applications. The request of a task may have more than one back-up resources to meet demand in the ad hoc cluster computing system. How to choose the best resources available to users is the core of system. This paper proposes a kind of genetic algorithm for task scheduling in ad hoc cluster computing in view of mathematical description and analysis of the capacity of the nodes, network resources and energy of the nodes. The experimental results show that the algorithm can make full use of limited bandwidth resources and maximize the continual ability of network.