MicroRank: End-to-End Latency Issue Localization with Extended Spectrum Analysis in Microservice Environments
Authors:
Top Cited Papers
Open Access
- 19 April 2021
- conference paper
- Published by Association for Computing Machinery (ACM)
- p. 3087-3098
Abstract
With the advantages of flexible scalability and fast delivery, microservice has become a popular software architecture in the modern IT industry. However, the explosion in the number of service instances and complex dependencies make the troubleshooting extremely challenging in microservice environments. To help understand and troubleshoot a microservice system, the end-to-end tracing technology has been widely applied to capture the execution path of each request. Nevertheless, the tracing data are not fully leveraged by cloud and application providers when conducting latency issue localization in the microservice environment. This paper proposes a novel system, named MicroRank, which analyzes clues provided by normal and abnormal traces to locate root causes of latency issues. Once a latency issue is detected by the Anomaly Detector in MicroRank, the cause localization procedure is triggered. MicroRank first distinguishs which traces are abnormal. Then, MicroRank’s PageRank Scorer module uses the abnormal and normal trace information as its input and differentials the importance of different traces to extended spectrum techniques . Finally, the spectrum techniques can calculate the ranking list based on the weighted spectrum information from PageRank Scorer to locate root causes more effectively. The experimental evaluations on a widely-used open-source system and a production system show that MicroRank achieves excellent results not only in one root cause situation but also in two issues that happen at the same time. Moreover, MicroRank makes 6% to 22% improvement in recall in localizing root causes compared to current state-of-the-art methods.Keywords
This publication has 26 references indexed in Scilit:
- Practitioners' expectations on automated fault localization2016
- Diagnosing Performance Variations by Comparing Multi-Level Execution TracesIEEE Transactions on Parallel and Distributed Systems, 2016
- An Unsupervised Spatiotemporal Graphical Modeling Approach to Anomaly Detection in Distributed CPS2016
- Dynamic time warping based causality analysis for root-cause diagnosis of nonstationary fault processesIFAC-PapersOnLine, 2015
- CauseInfer: Automatic and distributed performance diagnosis with hierarchical causality graph in large distributed systems2014
- Cloud SLAsACM SIGOPS Operating Systems Review, 2012
- A model for spectra-based software diagnosisACM Transactions on Software Engineering and Methodology, 2011
- Detailed diagnosis in enterprise networksACM SIGCOMM Computer Communication Review, 2009
- A Crosstab-based Statistical Method for Effective Fault Localization2008
- Scaling personalized web search2003