基于贪心算法的容器云资源低能耗部署方法

Low energy consumption deployment method of container cloud resources based on greedy algorithm

摘要:

将容器云平台资源整体能耗最低作为目标，设计基于贪心算法的容器云资源低能耗部署方法。在物理主机与虚拟机对应、虚拟机与容器对应等约束条件下，结合静态和动态两个部分构建容器云资源能耗模型。通过资源虚拟化与去除冗余两个步骤，得到容器云资源的整合结果。检测物理机负载状态，确定虚拟机迁移源物理机和目标物理机，利用贪心算法均衡调度容器云资源负载，最终通过容器云资源编排重组，实现容器云资源低能耗部署。通过与传统部署方法的对比得出结论：在优化设计部署方法下，容器云资源的利用率和负载均衡度得到明显提升，能量损耗明显下降。

Abstract:

Take the lowest overall energy consumption of container cloud platform resources as the goal to design a low-energy deployment method of container cloud resources based on greedy algorithm. Under the constraints of the correspondence between the physical host and the virtual machine and the correspondence between the virtual machine and the container, the container cloud resource energy consumption model is constructed by combining the static and dynamic parts. Through the two steps of resource virtualization and redundancy removal, the integration results of container cloud resources are obtained. Detect the load status of physical machines, determine the source and target physical machines of virtual machine migration, use greedy algorithm to balance and schedule the load of container cloud resources, and finally realize the low-energy deployment of container cloud resources through the arrangement and reorganization of container cloud resources. By comparing with the traditional deployment methods, it is concluded that under the optimized design deployment method, the utilization and load balance of container cloud resources are significantly improved, and the energy loss is reduced.