A New Replica Selection Strategy Based on Combination Algorithm
Zhengwei Guo, Ning Li, Hui Li, Nan Shi · 2010
In the data grid environment, when users access to files, how to select the best site to obtain files from multiple replicas and reach the highest QOS(quality of service) in the cost of same price is a problem that need to be studied urgently, that is replica selection. In this paper, it proposes a new combination algorithm based on genetic algorithm and ant algorithm, which not only solves the inefficient disadvantage of genetic algorithm, but also solves the problem of solving problems slowly in the early period in ant algorithm. This strategy is analyzed and simulated by data grid simulator OptorSim, and the results show that this new algorithm can be effective in reducing data access latency, improve data access speed, speed up job completion time, and achieve load balancing.