高级检索

  • ISSN 1006-3080
  • CN 31-1691/TQ

一种基于主副备份的微服务可靠性建模与分析方法

刘政 虞慧群 范贵生

刘政, 虞慧群, 范贵生. 一种基于主副备份的微服务可靠性建模与分析方法[J]. 华东理工大学学报(自然科学版). doi: 10.14135/j.cnki.1006-3080.20210921001
引用本文: 刘政, 虞慧群, 范贵生. 一种基于主副备份的微服务可靠性建模与分析方法[J]. 华东理工大学学报(自然科学版). doi: 10.14135/j.cnki.1006-3080.20210921001
LIU Zheng, YU Huiqun, FAN Guisheng. A Method to Model and Analyze Microservice Reliability Based on Primary-Backup Replication[J]. Journal of East China University of Science and Technology. doi: 10.14135/j.cnki.1006-3080.20210921001
Citation: LIU Zheng, YU Huiqun, FAN Guisheng. A Method to Model and Analyze Microservice Reliability Based on Primary-Backup Replication[J]. Journal of East China University of Science and Technology. doi: 10.14135/j.cnki.1006-3080.20210921001

一种基于主副备份的微服务可靠性建模与分析方法

doi: 10.14135/j.cnki.1006-3080.20210921001
基金项目: 上海市自然科学基金(21ZR1416300)
详细信息
    作者简介:

    刘政:刘 政(1989-),男,江苏人,博士生,主要研究方向为软件工程、云应用、微服务、可靠性设计、微服务组合、工作流、容错设计。E-mail:y10180300@mail.ecust.edu.cn

    通讯作者:

    虞慧群,E-mail: yhq@ecust.edu.cn

  • 中图分类号: TP393

A Method to Model and Analyze Microservice Reliability Based on Primary-Backup Replication

  • 摘要: 微服务架构(MSA)将云应用构建为独立的组件,通过解耦来提高软件更新的灵活性和敏捷性,然而微服务在时延和可靠性方面仍面临许多挑战。本文使用谓词Petri网(PrT网)对基于事件总线的微服务组合进行建模,建立微服务之间的逻辑关系。采用主版本和备份部署微服务实例,提高了微服务组合整体的可靠性。基于PrT网的相关理论,验证了微服务可靠性模型和PB副本部署策略的正确性。将微服务执行的保证率作为可靠性的参数,通过实验分析验证了本文方法的有效性。

     

  • 图  1  基于事件总线的微服务组合建模

    Figure  1.  Microservice composition modeling based on event bus

    图  2  基于事件总线的微服务组合HMCN 网建模

    Figure  2.  Modeling of microservice composition HMCN network based on event bus

    图  3  备份执行

    Figure  3.  Backup execution

    图  4  PB备份主动和被动执行模式的HMCN 网建模

    Figure  4.  HMCN network modeling of PB backup active and passive execution modes

    图  5  备份执行模式页面

    Figure  5.  Backup execution mode page

    图  6  两个任务部署在同一宿主机

    Figure  6.  Two tasks for the same hosts

    图  7  不同宿主机的两个任务

    Figure  7.  Two tasks for the different hosts

    图  8  PB被部署在不同容器内的HMCN网页面

    Figure  8.  HMCN page with PB deployed in the different containers

    图  9  PB被部署在同一容器内的HMCN网页面

    Figure  9.  HMCN page with PB deployed in the same containers

    图  10  基于PB副本部署的容器HMCN 网建模

    Figure  10.  Container HMCN network modeling based on PB replica deployment

    图  11  PB副本部署策略的可靠性

    Figure  11.  Reliability of PB replica deployment strategy

    图  12  PB副本部署策略的平均执行时间

    Figure  12.  Average execution time of PB replica deployment policy

  • [1] DAI F, MO Q, QIANG Z, et al. A Choreography analysis approach for microservice composition in cyber-physical-social systems[J]. IEEE Access, 2020, 8: 53215-53222. doi: 10.1109/ACCESS.2020.2980891
    [2] FAY A, BARTH M, STUTZ A, et al. Choreographies in microservice-based automation architectures - next level of flexibility for industrial cyber-physical systems[C]// 3rd IEEE International Conference on Industrial Cyber-Physical Systems. Finland: IEEE, 2020: pp. 411-416.
    [3] BARCHRAS M, KONTOGIANNIS K . Goal modelling meets service choreography: A graph transformation approach[C]// 2020 IEEE 24th International Enterprise Distributed Object Computing Conference (EDOC). USA: IEEE, 2020: 30-39.
    [4] ZHENG S, TILEVICH E . Equivalence-enhanced microservice workflow orchestration to efficiently increase reliability[C]// 2019 IEEE International Conference on Web Services (ICWS). Italy: IEEE, 2019: 426-433.
    [5] LI X, KAI L, PANG X, et al. An orchestration based cloud auto-healing service framework[C]// IEEE International Conference on Edge Computing. USA: IEEE, 2017: 190-193.
    [6] WU H, HUA X, LI Z, et al. Resource and instance hour minimization for deadline constrained DAG applications using computer clouds[J]. IEEE Transactions on Parallel and Distributed Systems, 2016, 27(3): 885-899. doi: 10.1109/TPDS.2015.2411257
    [7] MALAWSKI M, GIDEON J, Deelman E, et al. Algorithms for cost- and deadline-constrained provisioning for scientific workflow ensembles in IaaS clouds[J]. Future Generation Computer Systems, 2015, 48: 1-8. doi: 10.1016/j.future.2015.01.004
    [8] KISS T, KACSUK P, KOVACS J, et al. MiCADO–Microservice-based cloud application-level dynamic orchestrator[J]. Future Generation Computer Systems, 2017, 94: 937-946.
    [9] LIU Z, FAN G S, YU H Q, et al. An approach to modeling and analyzing reliability for microservice-oriented cloud applications[J/OL]. Wireless Communications and Mobile Computing, (2021-08-26) [2021-09-20]. https://www.hindawi.com/journals/wcmc/2021.
    [10] JIN W P, LIU Y Q, WU Q, . Fault-tolerant task scheduling in multiprocessor systems based on primary-backup scheme[C]// 2010 3rd International Symposium on Systems and Control in Aeronautics and Astronautics. USA: IEEE, 2010: 670-675.
    [11] XIE G, ZENG G, LI R, et al. Quantitative fault-tolerance for reliable workflows on heterogeneous iaas clouds[J]. IEEE Transactions on Cloud Computing, 2017, 8(4): 1223-1236.
    [12] XIAO W, XUE J, MIAO Y, et al. Distributed graph computation meets machine learning[J]. IEEE Transactions on Parallel and Distributed Systems, 2020, 31(7): 1588-1604. doi: 10.1109/TPDS.2020.2970047
    [13] AKSAKALLI I K, ELIK T, CAN A B, et al. Deployment and communication patterns in microservice architectures: A systematic literature review[J]. Journal of Systems and Software, 2021, 180(1): 111014.
    [14] LYU Z, WEI H, BAI X, et al. Microservice-based architecture for an energy management system[J]. IEEE Systems Journal, 2020, 14(4): 5061-5072.
    [15] CORTELLESSA V, DD POMPEO, ERAMO R, et al. A model-driven approach for continuous performance engineering in microservice-based systems[J]. Journal of Systems and Software, 2021(10): 111084.
  • 加载中
图(12)
计量
  • 文章访问数:  33
  • HTML全文浏览量:  39
  • PDF下载量:  2
  • 被引次数: 0
出版历程
  • 收稿日期:  2021-09-21
  • 录用日期:  2021-12-07
  • 网络出版日期:  2022-04-12

目录

    /

    返回文章
    返回