基于QoS和效用的云计算资源调度模型

发布时间：2018-08-14 10:45

【摘要】：针对已有的云计算环境下资源调度模型往往仅考虑任务执行跨度而忽略了能提高用户满意度的效用因素,为此,提出了一种基于用户需求QoS和最大化效用为目标的云计算资源调度模型;首先给出了云计算资源调度的框架和数学模型,并将满足用户预算和最迟完完工时间约束的总效用函数作为优化目标;然后对所有任务构造决策矩阵和归一化处理,并采用拉格朗日松弛求取属性权重向量,从而构造出最终的基于多维属性的总效用函数;最后,定义了基于QoS和以最大化效用为目标的云计算资源调度算法;采用CloudSim工具进行仿真实验,结果表明:文中方法能获得最优的资源调度方案,且与其它方法相比,具有较少的平均执行时间和执行花费,具有较大的优越性。
[Abstract]:In view of the existing resource scheduling model in cloud computing environment, only the task span is considered and the utility factors that can improve the user satisfaction are ignored. This paper proposes a cloud computing resource scheduling model based on user demand QoS and maximization of utility. Firstly, the framework and mathematical model of cloud computing resource scheduling are given. The total utility function, which satisfies the user's budget and the deadline of completion time, is taken as the optimization objective, and then the decision matrix and normalized processing are constructed for all tasks, and the attribute weight vector is obtained by Lagrange relaxation. Finally, the paper defines the resource scheduling algorithm based on QoS and maximization of utility, and uses CloudSim tools to carry out the simulation experiment, and then constructs the final utility function based on multidimensional attributes, finally, defines the scheduling algorithm of cloud computing resources based on QoS and the goal of maximizing utility. The results show that the proposed method can obtain the optimal resource scheduling scheme, and compared with other methods, it has less average execution time and execution cost, and has more advantages.
【作者单位】：安阳工学院飞行学院;安阳工学院北校区计算机系;
【基金】：国家自然科学基金河南人才培养联合基金项目(U1204613)
【分类号】：TP393.02

【参考文献】