This paper proposed a large-scale resource scheduling method based on improved genetic algorithm combined with secondary coding. This method solved the problem that traditional genetic algorithm could not meet the resource scheduling problem in large-scale cloud computing environment under multi-user. In the selective replication phase, it used a dual fitness function based on minimum task completion time and matching degree to screen the populations by double criteria. Next, it adaptively optimized the cross-mutation probability of the algorithm, and further improved its adaptive ability, which ensured that the algorithm converged to the optimal solution as soon as possible. The introduced convergence termination condition ensured that the algorithm jumped out of the loop as soon as possible. Finally, it used the CloudSim platform to analyze the improved genetic algorithm(IGA). It shows that the improved genetic algorithm can be well applied to large-scale resource scheduling, and the result is better than the comparison algorithm.

[2020-08-05] Printed Article /doc/2019.02.0060/v2