Exploring Nonintrusive Measurements of Spatio‐temporal Portrait of Microservices
Tao Zeng,Zichen Xu,Dan Wu,Xiaoling Li,Biyong Liu,Haichuan Hu,Shuang Tan,Yusong Tan,Chenren Xu,Christopher Stewart,Qihe Zhou,Ye Cao
DOI: https://doi.org/10.1002/spe.3234
2024-01-01
Abstract:As cloud native technology advances, the scale and complexity of applications built on microservice architecture continue to expand, leading to increasingly intricate differences between software within the same application. Microservice applications, offering high flexibility, are deployed in data centers as black boxes from the users' perspective, leaving them with no insight into the orchestration of cloud service providers. Consequently, users face challenges in promptly recognizing performance imbalances within their deployed applications. Meanwhile, cloud service providers may cut costs by offering a mix of qualified and unqualified services, potentially deceiving users. To enhance the understanding of microservice application organization, we propose a non-intrusive measurement framework, termed NMPI. NMPI facilitates rapid identification of microservice application defects, offering insights into cloud services and detecting fraudulent behavior in microservice-based applications. We model microservice applications using a queue analysis-based approach and filter the dominant frequency components of average response time signals by employing k-means on the fast fourier transform (FFT). Our model constructs a library of performance portraits for various software, with these portraits resembling human fingerprints that carry and mark the software's internal information. Utilizing a two-tier microservices-based application incorporating a database as a case study allows us to demonstrate the effectiveness of NMPI. Our experimental results show that NMPI can produce differentiable profiles of data service performance portraits across a diverse and extensive range of workloads, enabling the identification of software types and the analysis of performance conditions.