• 中国计算机学会会刊
  • 中国科技核心期刊
  • 中文核心期刊

计算机工程与科学 ›› 2020, Vol. 42 ›› Issue (07): 1151-1157.

• 高性能计算 • 上一篇    下一篇

基于Kubemark的微服务性能仿真测试研究

雷擎   

  1. (对外经济贸易大学信息学院,北京 100029)
  • 收稿日期:2019-07-13 修回日期:2020-01-13 接受日期:2020-07-25 出版日期:2020-07-25 发布日期:2020-07-25
  • 作者简介:雷擎 (1969),女,重庆人,博士,讲师,研究方向为模式识别、自然语言处理和高性能计算。 E-mail:leiqing@uibe.edu.cn
  • 基金资助:
    北京市社会科学基金(18GLB021);对外经济贸易大学中央高校基本科研业务费专项基金(17YB20)

Microservice performance simulation test based on Kubemark

LEI Qing   

  1. (School of Information Technology & Management,University of International Business and Economics,Beijing 100029,China)
  • Received:2019-07-13 Revised:2020-01-13 Accepted:2020-07-25 Online:2020-07-25 Published:2020-07-25
  • About author:LEI Qing ,born in 1969,PhD,lecturer,her research interests include pattern recognition,natural language processing,and high performance computing.

摘要: 虚拟化技术加速了微服务架构上应用程序的扩展,随着这些应用程序复杂性不断增加,系统实际的性能可能会与预期存在很大差异,因此微服务性能测试机制成为了学者们开始探索的课题。借鉴Web服务质量的测试方法和评价标准,在实验过程中采用了仿真测试方法,通过Kubemark工具基于Kubernetes平台对微服务系统的性能进行测试研究,并根据RFC 2679标准的p百分位数指标对测试结果进行了分析。实验结果表明,微服务性能受所负载微服务类型的影响明显,仿真测试是微服务性能测试的有效研究方法。


关键词: 微服务, 微服务性能, Kubernetes, Kubemark, 云服务

Abstract: Virtualization technology has greatly accelerated the expansion of applications on the microservice architecture. As the complexity of these applications continues to increase, microservice performance may be significantly different from the expected. Therefore, the methods and evaluation criteria of microservice performance testing have become a topic that scholars have begun to explore. This paper draws on the testing methods and evaluation standards of Web service quality, and uses the simulation testing methods in the experiments. The Kubemark tool is used to test the performance of the microservice system on the Kubernetes platform, and the p-percentile indicator in RFC 2679 standard is used to analyze the results. Experimental results show that the performance of microservices is significantly affected by the type of load microservices, and simulation testing is an effective research method for microservices performance testing.

Key words: microservice, microservice performance, Kubernetes, Kubemark, cloud service

中图分类号: