OpenStack云平台的高可用设计与实现

发布时间：2019-05-18 21:14

【摘要】：实现高可用性是OpenStack云计算管理平台研究的重要问题之一。针对OpenStack云计算管理平台的相关服务组件运行在单节点上易导致单点故障(SPoF)的问题,结合现有多种系统高可用性解决方案,提出一种基于Pacemaker+Corosync+HAProxy+Ceph的解决方案以实现OpenStack云计算管理平台的高可用。该方案将Active-Active的双活模式、Active-Passive的主备模式及集群技术3种高可用设计模式融合在一起,通过软硬件冗余和服务实例故障转移等方式实现OpenStack云计算管理平台的高可用性。实验证明,在少量节点或链路中断的情况下OpenStack云计算管理平台仍然能够稳定运行,该高可用性方案具有可行性。
[Abstract]:Realizing high availability is one of the important problems in the research of OpenStack cloud computing management platform. Aiming at the problem that the related service components of OpenStack cloud computing management platform run on a single node, which can easily lead to a single point of failure (SPoF), combined with a variety of existing system high availability solutions, A solution based on Pacemaker Corosync HAProxy Ceph is proposed to realize the high availability of OpenStack cloud computing management platform. In this scheme, the dual active mode of Active-Active, the main standby mode of Active-Passive and the cluster technology are combined to realize the high availability of OpenStack cloud computing management platform through hardware and software redundancy and service instance fault transfer. The experimental results show that the OpenStack cloud computing management platform can still run stably in the case of a small number of nodes or link outages, and the high availability scheme is feasible.
【作者单位】：西南石油大学计算机科学学院;
【基金】：四川省教育厅专项基金018-数字油田监控(020402000018)资助
【分类号】：TP393.09

【相似文献】