首页|基于OpenStack的高可用城轨云平台:设计、实现与可用性分析

基于OpenStack的高可用城轨云平台:设计、实现与可用性分析

扫码查看
轨道交通系统对于可靠性具有特殊要求,构建高可用轨道交通私有云计算平台,已成为城轨交通的热门研究领域.设计并实现高可用云管系统与高可用云上虚拟机,能有力推动更多的轨道交通业务向云平台迁移,提升资源利用率.首先,介绍轨道交通的云计算建设概况及云平台容错方法;其次,详细描述利用开源OpenStack技术,实现高可用云管平台与高可用虚拟机的方法,利用故障树工具分析云管系统与虚机的失效模式之后,用马尔科夫过程对云管系统与虚机的可靠性进行建模分析;最后,利用设计的高可用云平台,部署完整的基于通信的列车控制系统中的地面业务,并对虚拟机两项可用性的关键指标进行测试,根据测试结果,获得模型的参数,进而计算得到整个私有云计算平台的可用性指标.本研究将为城市轨道交通的高可用云计算平台建设提供有力的理论基础与实践经验.
A High-availability Urban Rail Cloud Platform Based on OpenStack:Design,Implementation and Availability Analysis
Rail transit systems have special requirements for reliability.Building a high availability(HA)rail transit private cloud computing platform has become a hot research field for urban rail transit.Designing and implementing a highly available cloud management system and virtual machines(VMs)on the cloud can effectively promote the migra-tion of more rail transit services to the cloud computing platform and improve the resource utilization rate of the rail tran-sit.Firstly,the general situation of cloud computing construction in rail transit was introduced,with emphasis on the fault-tolerant method of cloud computing platform.Subsequently,the specific methods of using the open-source Open-Stack technology to realize the HA cloud management platform and the HA VMs were described.Based on the analysis of the failure modes of the cloud management system and the VMs using the fault tree tool,the reliability model of the cloud management system and the VMs was established by the Markov process.Finally,the ground service in a complete Communication Based Train Control(CBTC)system was deployed by using the designed HA cloud computing platform,with two key indicators being tested to reflect VM's availability.According to the test results,the parameters of the reliability model were obtained to further calculate the availability indexes of the entire private cloud computing platform.The research will provide a strong theoretical basis and practical experience for the construction of a HA cloud computing platform for urban rail transit.

urban rail transitcloud computingOpenStackvirtual machinehigh availabilityreliability

朱力、李子恒、唐涛、王悉

展开 >

北京交通大学轨道交通信号控制与安全国家重点实验室,北京 100044

城市轨道交通 云计算 OpenStack 虚拟机 高可用 可靠性

国家重点研发计划国家自然科学基金北京市自然科学基金国家重点实验室自主课题

2020YFB131330161973026L201002RCS2021ZZ005

2024

铁道学报
中国铁道学会

铁道学报

CSTPCD北大核心
影响因子:0.9
ISSN:1001-8360
年,卷(期):2024.46(2)
  • 16