一种基于图模型的可信云资源调度算法

Graph model based trustworthy resource scheduling algorithm in cloud environment

根据并行任务及胖树形云系统的结构特点,综合考虑了云计算环境下资源节点与通信链路的可靠性问题,构建了基于任务执行行为的云系统可靠性度量模型。提出了一种基于图的可信云资源调度模型,将云任务可信资源需求与云资源动态供给的最优匹配问题转换成最小费用最大流图的构造和求解问题。针对图的求解计算复杂度较高的问题,结合商空间理论将初始网络转化为规模较小的商网络进行求解。仿真实验结果表明,该算法能够有效提高云任务执行的成功率,降低基于图模型的资源调度算法延迟。

The characteristics of the graph structure of parallel task and fat tree cloud system are analyzed. Then,by combining cloud tasks trust requirement for resource nodes together w ith the trust requirement for communication links,an execution-behavior based reliability measurement model is established. Then,w e present a graph based trustw orthy cloud resource scheduling model. It is easy to express taskstrust scheduling requirement and resource supply on a unified w ay,w hich can be transformed to a minimum cost maximum flow netw ork construction and solution problem. Finally,in consideration of reducing computational complexity,algorithms are applied on the quotient netw ork to approximately obtain the minimum cost flow of original netw ork. The simulation experimental results show that the proposed algorithms can effectively improve the success rate of cloud tasks execution and significantly reduce runtime of flow netw ork.