Simulated Annealing Based On Ant Colony Algorithm For Solving The Grid Of Task Scheduling Problem

With the rapid changes in technology, the computer network played an increasingly important role in our lives. I can say that the essential point of Off-site resource sharing network brought us a lot of convenience and off-site resources on how to better share gradually become a new research direction. Grid task scheduling is to discuss the problem, which design a good resource scheduling algorithm is the main purpose of this study.The ant colony algorithm is a heuristic algorithm. It is a positive feedback mechanism proposed by the Italian scholar Dorigo. M and it is inspired in the path of the ants foraging path finding. In this process, ants released information has always been to play the role of guide other ants. Behind the ants will be based on how much the previous ants release pheromones to explore the path. But the basic ant colony algorithm is a local optimum early convergence phenomenon, so in some cases they will not necessarily find the optimal path. The instances of the common application are mostly to be improved. In the text, we will propose an ant colony algorithm based on simulated annealing to solve the grid task scheduling problem.This article firstly discussed the research background of domestic and foreign, and described the status of grid task scheduling with ant colony algorithm, where the need to be improved. Then it introduced the basic concepts, principles, and several variations of ant colony algorithm, and described grid tasks in detail, including the concepts, characteristics and significance of the grid, and introduced a detailed description of several architecture models and several scheduling algorithms, one of which proposed a new genetic algorithm for task scheduling in grid, which combined elitist selection and selection of the roulette plate to improve the efficiency of the algorithm, and used agreement hybridization to avoid the loss of the gene as well as to improve the convergence rate.The article also discussed the general ant colony algorithm in detail to solve the grid task scheduling application, but when the ant colony algorithm explores the path will lead to local optima, this path is not necessarily the global optimum and may be there exists other global optimum path, however, this moment the ants might have found a good path and will not continue to find the global optimal path. Annealing algorithm is a kind of local search strategy by means of a certain probability to accept a lower solution and to avoid local optimal. Therefore the idea of simulated annealing added to this article in the basic of ant colony algorithm, and their combination well solved the contradiction of such a local optimum. The simulation results also show that the improved algorithm can more effectively solve the problem of grid task scheduling.In short, the grid task scheduling, the grid task scheduling, whether in the level of theoretical research or not in real life application has very important significance. A good scheduling algorithm that plays a decisive role in the performance of grid task scheduling algorithm of this paper not only can solve the scheduling problem, but also exploit well field expansion of the ant colony algorithm.