基于SLA约束的IaaS虚拟机的多维资源放置算法

发布时间：2018-05-26 07:44

本文选题：IaaS层资源 + SLA　；参考：《微电子学与计算机》2017年10期

【摘要】：针对IaaS层资源放置中的虚机放置问题,提出了IaaS云资源放置SLA的服务质量指标模型和基于QoS的多维度差分进化放置算法。算法采用差分进化算法结合非支配排序思想,通过对非支配排序在排序方式以及子代筛选上进行优化,实现对IaaS的资源放置时兼顾多种类资源约束以及其他服务质量指标约束,从而获得更合理的多类型资源放置结果.实验结果表明本文提出的多维资源虚机放置算法搜索速度快,多维资源放置结果产生的违约率更低,在QoS指标上具备更优的性能.
[Abstract]:Aiming at the problem of virtual machine placement in IaaS layer resource placement, the QoS index model of IaaS cloud resource placement SLA and the multi-dimensional differential evolutionary placement algorithm based on QoS are proposed. The algorithm adopts the differential evolution algorithm combined with the idea of non-dominated sorting, and optimizes the sorting mode and the selection of offspring in order to realize the resource placement of IaaS with multi-resource constraints and other quality of service constraints. In order to obtain more reasonable results of multi-type resource placement. The experimental results show that the multi-dimensional resource placement algorithm proposed in this paper has a fast searching speed, a lower default rate and a better performance on the QoS index.
【作者单位】：西安交通大学软件学院;西安交通大学电子与信息工程院;
【基金】：国家“八六三”基金项目(2014AA01A302)
【分类号】：TP18;TP302

【相似文献】