[1]郑杰生,谢彬瑜,吴广财,等.一种基于 Lasso 回归的微服务性能建模方法[J].计算机技术与发展,2020,30(12):216-220.[doi:10. 3969 / j. issn. 1673-629X. 2020. 12. 038]
　ZHENG Jie-sheng,XIE Bin-yu,WU Guang-cai,et al.A Lasso Regression Based Performance Modeling Method for Microservices[J].,2020,30(12):216-220.[doi:10. 3969 / j. issn. 1673-629X. 2020. 12. 038]
点击复制

一种基于 Lasso 回归的微服务性能建模方法()
分享到：

《计算机技术与发展》[ISSN:1006-6977/CN:61-1281/TN]

卷:
30
期数:
2020年12期
页码:
216-220
栏目:
应用开发研究
出版日期:
2020-12-10

文章信息/Info

Title:
A Lasso Regression Based Performance Modeling Method for Microservices
文章编号:
1673-629X(2020)12-0216-05
作者:
郑杰生1; 谢彬瑜1; 吴广财1; 陈 非1; 花 磊2
1. 广东电力信息科技有限公司,广东 广州 510000; 2. 苏州博纳讯动软件有限公司,江苏 苏州 215000
Author(s):
ZHENG Jie-sheng1; XIE Bin-yu1; WU Guang-cai1; CHEN Fei1; HUA Lei2
1. Guangdong Electric Power Information Technology Co. ,Ltd. ,Guangzhou 510000,China; 2. Suzhou Bona Xundong Software Co. ,Ltd. ,Suzhou 215000,China
关键词:
微服务; 性能建模; 容量规划; Lasso 回归; 云计算
Keywords:
microservices; performance modeling; capacity plan; Lasso regression; cloud computing
分类号:
TP312
DOI:
10. 3969 / j. issn. 1673-629X. 2020. 12. 038
摘要:
微服务技术广泛用于构建多样化的分布式软件,微服务的资源使用取决于其所实现的内部功能和处理的外部负载,负载突增会造成软件的性能衰减, 因此需要动态调整微服务的最大访问速率以保证其服务质量。 然而,在云计算环境下,软件的部署环境与应用类型具有多样性和复杂性,因而难以准确评估微服务处理请求的能力。 为了应对以上问题,提出一种基于 Lasso 回归的微服务性能建模方法。 首先将目标微服务放置在独立的 Docker 容器中,而后模拟生成微服务的外部负载并搜集其性能监测数据,进而基于 Lasso 回归建立资源与性能的关联模型以评估微服务的请求处理能力,从而实现微服务的细粒度灵活水平扩展。 最后实现了原型系统并进行典型微服务实验,结果表明系统及方法具有较低的预测误差,并能够为软件提供较好的性能保障。
Abstract:
Microservice technologies are widely used to develop various distributed software. The resource utilization of microservices depends on their implemented functions and processed external workloads. The surge of workloads can cause the performance degradation of software,so it is necessary to dynamically adjust the maximum access rate of each microservice to ensure its quality of service. However,since the deployment environment and application types of software are diverse and complex in cloud computing,it is difficult to accurately assess the service capability of each microservice to process requests. We propose a Lasso regression-based method of microservices’ performance modeling. First, the target microservice is placed in an independent Docker container,and then the external workload of the microservice is simulated to generate and its performance monitoring data is collected. Then,an association model of resources and performance is established based on Lasso regression to evaluate the request processing capacity of the microservice,so as to realize the fine-grained and flexible horizontal expansion of the microservice. The prototype system is implemented to evaluate the proposed method on a typical microservice. The experiment shows that the proposed method has low prediction error rate and can improve the overall performance.

相似文献/References:

[1]黄继杰,林昌年,杨选怀,等.一种支持私有云仿真的 HLA/ RTI 实现方法[J].计算机技术与发展,2019,29(09):164.[doi:10. 3969 / j. issn. 1673-629X. 2019. 09. 031]
　HUANG Ji-jie,LIN Chang-nian,YANG Xuan-huai,et al.An HLA/ RTI Implementation Supporting Private Cloud Simulation[J].,2019,29(12):164.[doi:10. 3969 / j. issn. 1673-629X. 2019. 09. 031]
[2]罗钦凯,倪成章.基于微服务的工作流技术在云管平台的应用[J].计算机技术与发展,2019,29(09):122.[doi:10. 3969 / j. issn. 1673-629X. 2019. 09. 024]
　LUO Qin-kai,NI Cheng-zhang.Application of Workflow Technology Based on Micro-service in Cloud Management Platform[J].,2019,29(12):122.[doi:10. 3969 / j. issn. 1673-629X. 2019. 09. 024]
[3]吴 磊,湛 健,宋丽华.微服务架构在智能家居网关系统中的应用研究[J].计算机技术与发展,2019,29(11):200.[doi:10. 3969 / j. issn. 1673-629X. 2019. 11. 040]
　WU Lei,ZHAN Jian,SONG Li-hua.Research of Application of Micro-service Architecture in Smart Home Gateway System[J].,2019,29(12):200.[doi:10. 3969 / j. issn. 1673-629X. 2019. 11. 040]
[4]姜 伟,潘邵芹.基于 SDN 的微服务负载均衡方案研究[J].计算机技术与发展,2020,30(02):23.[doi:10. 3969 / j. issn. 1673-629X. 2020. 02. 005]
　JIANG Wei,PAN Shao-qin.Research on Load Balance of Microservice Based on SDN[J].,2020,30(12):23.[doi:10. 3969 / j. issn. 1673-629X. 2020. 02. 005]
[5]郑杰生,谢彬瑜,吴广财,等.一种基于模式识别的微服务异常检测方法[J].计算机技术与发展,2020,30(11):123.[doi:10. 3969 / j. issn. 1673-629X. 2020. 11. 023]
　ZHENG Jie-sheng,XIE Bin-yu,WU Guang-cai,et al.An Anomaly Detection Approach for Microservices Based on Pattern Recognition[J].,2020,30(12):123.[doi:10. 3969 / j. issn. 1673-629X. 2020. 11. 023]
[6]耿晓利,张 芒,尹永宏.高并发高可用的分布式电商平台架构研究[J].计算机技术与发展,2021,31(02):111.[doi:10. 3969 / j. issn. 1673-629X. 2021. 02. 021]
　GENG Xiao-li,ZHANG Mang,YIN Yong-hong.Research on Distributed E-commerce Platform Architecture with High Concurrency and High Availability[J].,2021,31(12):111.[doi:10. 3969 / j. issn. 1673-629X. 2021. 02. 021]
[7]马勤政,徐中伟,梅 萌.基于 Kubernetes 的列控系统测试容器云平台设计[J].计算机技术与发展,2021,31(06):52.[doi:10. 3969 / j. issn. 1673-629X. 2021. 06. 010]
　MA Qin-zheng,XU Zhong-wei,MEI Meng.Design of Container Cloud Platform for Test of Train Control SystemBased on Kubernetes[J].,2021,31(12):52.[doi:10. 3969 / j. issn. 1673-629X. 2021. 06. 010]
[8]邵 瑛,徐 斌.一种通用的文本日志类信息分析评估模型[J].计算机技术与发展,2021,31(增刊):73.[doi:10. 3969 / j. issn. 1673-629X. 2021. S. 014]
　SHAO Ying,XU Bin.A Common Analysis and Evaluation Model of Text Log Information[J].,2021,31(12):73.[doi:10. 3969 / j. issn. 1673-629X. 2021. S. 014]
[9]邹春杰,赵学健,朱 涛,等.基于微服务架构的农产品溯源系统优化[J].计算机技术与发展,2022,32(01):147.[doi:10. 3969 / j. issn. 1673-629X. 2022. 01. 025]
　ZOU Chun-jie,ZHAO Xue-jian,ZHU Tao,et al.Optimization of Agricultural Product Traceability System Based onMicro-service Architecture[J].,2022,32(12):147.[doi:10. 3969 / j. issn. 1673-629X. 2022. 01. 025]
[10]鲍春林,宋丽华,余 航.基于节点分层和延迟敏感的服务放置策略[J].计算机技术与发展,2022,32(10):14.[doi:10. 3969 / j. issn. 1673-629X. 2022. 10. 003]
　BAO Chun-lin,SONG Li-hua,YU Hang.Service Placement Policies Based on Node Tiering and Latency Sensitivity[J].,2022,32(12):14.[doi:10. 3969 / j. issn. 1673-629X. 2022. 10. 003]

更新日期/Last Update: 2020-12-10