摘要
现有数据中心中时间感知型云计算应用的资源分配算法能耗较高,严重影响了数据中心的服务上限以及云服务商的经济效益,对此提出一种低能耗的云计算资源分配与调度优化算法.算法分为两个阶段:第一阶段,释放并更新请求集的服务器与链接的剩余容量,同时更新能量辅助图中相应的权重;第二阶段,将所有新到达请求按所需时间段以降序排列,为各请求分配资源;第三阶段,检查资源可用性,并基于能量辅助图采用最短权值路径选择算法为资源请求分配虚拟机与流量.基于思科真实设备参数的仿真实验结果表明,本文云计算资源分配与路由算法的能量效率与资源分配性能均优于其他算法.
The existing the resource allocation algorithms of time aware cloud computing applications in the data centre perform high energy consumption,and it significantly affect the ability to support more tenants and the economic benefits of cloud computing service providers,concerned at that problem a low power consumption resource allocation and scheduling algorithm is proposed.The algorithm consists of two phases：in the first phase,the residual capacities of services and links in the request are updated and released,and the corresponding weights in the energy graph are updated;in the second phase,the new arriving requests are ordered in descending order,and the resources are allocated to each request.And lastly,the availability of request is checked and the shortest weighted paths election algorithm is used to allocate the virtual machines and bandwidth for resource requests.Cisco device references based simulation experimental results show that the proposed cloud computing resource allocation and routing algorithm outperforms the other algorithms in energy efficiency and resource allocation performance.
出处
《西南师范大学学报（自然科学版）》
CAS
北大核心
2017年第7期46-53,共8页
Journal of Southwest China Normal University(Natural Science Edition)
基金
2017年广东省大数据分析与处理重点实验室开放基金资助项目(2017016)
揭阳职业技术学院2016年一般科研项目(2016JYCKY05)
关键词
云计算
时间感知应用
多租户
资源分配
虚拟机
带宽资源
cloud computing
time aware application
multi-tenancy technology
resource allocation
virtual machine
bandwidth resource
