基于OpenStack体系的多融合管道式服务云监控系统的研制

doi:10.3969/j.issn.1000-0801.2014.01.004

摘要/Abstract

摘要：

为了保证在云计算大集群规模下服务的稳定性和可靠性，针对目前云监控的需求，阐述了基于OpenStack工程体系结构研制的云监控系统，并详细论述了系统结构和关键技术点，包括所采用的工厂模式、Paste Deploy、管道、Eventlet协同程、面向对象的数据库SQL Alchemy等编程模式，使得系统具备了模块化、层次化、流水线化等结构特点，具备多模式融合、部署简易、高扩展性和二次开发简易等优势。测试结果表明，系统具备了功能服务的高稳定性、采集数据的高准确性以及与第三方客户端的易融合性，能有效地实现云计算平台的信息化监控。

关键词: 云监控, 管道结构, 协同程, 对象化数据编程, OpenStack

Abstract:

To deal with the increasing of computer clusters which the increasing needs of cloud computing brings and guarantee stability and reliability of the service, a cloud computing monitoring system based on OpenStack structure to satisfy the needs of cloud computing monitor was designed. The system structure and key technologies were described, including factory model, Paste Deploy, pipeline, Eventlent green thread and object-oriented SQL Alchemy model, so the system has modularity, layering and pipeline. The advantage of the system includes multi-function integrating, easy deploy, high expansibility and redeveloping easily. The experimental result proves that the system has high stability, accurate data acquisition and high integration with the third client, and it can implement cloud computing monitoring informatization more effectively.

Key words: cloud computing monitoring, pipeline structure, eventlent green thread, SQL Alchemy, OpenStack

李磊,金连文,黎德生,唐万万. 基于OpenStack体系的多融合管道式服务云监控系统的研制[J]. 电信科学, 2014, 30(1): 24-30.

Lei Li,Lianwen Jin,Desheng Li,Wanwan Tang. Research on Multi-Function Pipeline Integrate Cloud Computing Monitoring System Based on OpenStack Structure[J]. Telecommunications Science, 2014, 30(1): 24-30.

图/表 12

图1

表1

图2

图3

表2

图4

图5

表3

图6

图7

图8

表4

参考文献 15

1	沈青，董波，肖德宝．基于服务器集群的云监控系统设计与实现．计算机工程与科学 2012,34（10）: 73～77
2	Mikyung K , Dong-In K , Stephen P . et al. Design and development of a run-time monitor for multi-core architectures in cloud computing. Sensors, 2011, 11 （4）: 3595～3610
3	Shao J , Wei H , Wang Q X , et al. A runtime model based monitoring approach for cloud. Proceedings of IEEE 3rd International Conference on Cloud Computing, Florida, 2010: 313～320
4	赵海川，吴江，王鹏等．面向电信业务的基础设施云监控系统．西北大学学报（自然科学版） 2012,42（5）: 729～734
5	Ruan L , Peng J B , Xiao L M , et al. Distributed virtual machine monitor for distributed cloud computing nodes integration. Lecture Notes in Computer Science, 2013（7861）: 23～31
6	DeChaves S A , Uriarte R B , Westphall C B . Toward an architecture for monitoring private clouds. Lecture Notes in Computer Science, IEEE, 2011（49）: 103～137
	7,2012
8	Rick C . Essential SQL Alchemy. OˊReilly Media Inc, 2008
9	Stacklab:documentation:OpenStack-hacker. ,2013
10	Thamarai S , Somasundaram , Kannan G . Cloud monitoring and discovery service（CMDS）for IaaS resources. Proceedings of the Third International Conference on Advanced, India, 2011: 340～345
11	董波，沈青，肖德宝．云计算集群服务器系统监控方法的研究．计算机工程与科学 2010,34（10）: 69～72
12	12 Get CPU usage with get system times．,2011
13	Paste Deployment．,2012
14	Zhao W B , Melliar-Smith P M , Moser L E . Fault tolerance middleware for cloud computing. Proceedings of IEEE the 3rd International Conference on Cloud Computing, India, 2010: 67～74
15	Curl. ,2011

数据采集类型	采集内容	获取方式
物理主机运行信息	物理主机各项运行信息	Python库
虚拟云主机运行信息	CPU、内存、磁盘I/O	虚拟化管理库libvirt
虚拟云主机运行统计信息	磁盘使用量、网络传输量	虚拟化管理库libvirt

协同程名称	采集内容	采集模块
PHY performance task	物理主机各项运行信息	Python-psutil
instance performance task	CPU、内存、磁盘I/O	Python-libvit
instance statistic task	磁盘使用量、网络传输量	Python-libvit
heartbeat task	采集点服务健康心跳信息	无

curl指令格式	指令说明
PHY performance task	Python-psutil
instance performance task	Python-libvit
instance statistic task	Python-libvit

采集信息类型（每小时平均值）	系统采集信息	top指令
CPU占用率	15%	13%
内存使用量	512 MB	505 MB
硬盘I/O压力（写入）	43 555 667 byte	43 567 889 byte
硬盘I/O压力（读出）	29 968 829 byte	29 668 552 byte
网卡流量	32 263 byte	32 568 byte

[1]	冯伟斌,张云勇,王志军. 支持不同存储方式的OpenStack云平台设计与实现[J]. 电信科学, 2016, 32(4): 141-145.
[2]	陈天,陈楠,黄志兰,樊勇兵,赖培源. 基于OpenStack的异构混合云解决方案[J]. 电信科学, 2015, 31(7): 132-138.
[3]	庄子睿,王敬宇,徐童. 面向租户的虚拟机定制化组网技术[J]. 电信科学, 2015, 31(10): 144-150.
[4]	张园,张云勇,房秉毅. 一种基于工作流的云系统自动化部署架构[J]. 电信科学, 2014, 30(11): 14-21.
[5]	姜毅,王伟军,曹丽,刘凯,陈桂强. 基于开源软件的私有云计算平台构建[J]. 电信科学, 2013, 29(1): 68-75.
[6]	李小宁,李磊,金连文,黎德生. 基于OpenStack构建私有云计算平台[J]. 电信科学, 2012, 28(9): 1-8.
[7]	赵淦森,王维栋,孙伟,季统凯. 云计算平台生存性研究[J]. 电信科学, 2011, 27(9): 52-59.