快速降阶匈牙利算法的云计算任务分配模型

Task assignment model in cloud computing based on Hungary algorithm of faster reduced order

为了提高云计算任务分配效率,在标准匈牙利算法的基础上,提出一种快速降阶优化算法的云计算任务分配模型.为实现快速求解全局任务分配问题,快速降阶算法不断排除已确定的分配方案对应的代价矩阵元素,从而快速降低矩阵的阶次.并可根据成本矩阵规模将矩阵分解成多个矩阵,使得该算法在任务和计算机不对等的情况下同样适用.论文最后的仿真结果表明,快速降阶匈牙利算法计算耗时远远小于匈牙利算法,并能有效提高计算机的利用率.

In order to improve the efficiency of task allocation in cloud computing, on the basis of standard Hungary algorithm, this paper puts forward a task allocation model in cloud computing using fast reduced order optimization algorithm. To solve all task allocation problem fast, the algorithm constantly rules out corresponding matrix elements in the allocation plan, and thus the order of the matrix is quickly reduced. And the matrix is decomposed into multiple matrixes according to the size of cost matrix, which makes the algorithm also applicable in the uncoordinated case of tasks and computer. The final simulated results of the paper show that Hungary algorithm in the fast reduced order costs far less time than that of Hungarian algorithm, and the former improves the utilization of the computer.