A Root Cause Analysis Framework for Microservice Systems with Multimodal Data

doi:10.12142/ZTECOM.202504012

ZTE Communications ›› 2025, Vol. 23 ›› Issue (4): 110-119.DOI: 10.12142/ZTECOM.202504012

• Research Papers • 上一篇下一篇

收稿日期:2024-01-26 出版日期:2025-12-25 发布日期:2025-12-22

A Root Cause Analysis Framework for Microservice Systems with Multimodal Data

LI Yingke¹, HAN Jing²(), SUN Yongqian¹, SHI Binpeng¹, GONG Zican²

^1.Nankai University, Tianjin 300071, China
^2.ZTE Corporation, Shenzhen 518057, China

Received:2024-01-26 Online:2025-12-25 Published:2025-12-22
About author:LI Yingke received her BS degree in software engineering from the School of Information Engineering, Minzu University of China in 2018. She is currently pursuing her master’s degree at the College of Software, Nankai University, China. Her research interests include anomaly detection and failure diagnosis.
HAN Jing (han.jing28@zte.com.cn) received her master’s degree from Nanjing University of Aeronautics and Astronautics, China. She has been with ZTE Corporation since 2000. She had been engaged in 3G/4G key technologies, from 2000 to 2016, and has become a technical director responsible for intelligent operation of cloud platforms and wireless networks since 2016. Her research interests include machine learning, data mining, and signal processing.
SUN Yongqian received his BS degree in statistical specialty from Northwestern Polytechnical University, China in 2012, and PhD degree in computer science from Tsinghua University, China in 2018. He is currently an assistant professor with the College of Software, Nankai University, China. His research focuses on anomaly detection, root cause analysis, and failure diagnosis in service management.
SHI Binpeng received his BE degree in software engineering from the College of Software, Nankai University, China in 2023, where he is currently pursuing his master’s degree. His research interests include anomaly detection and failure diagnosis.
GONG Zican received his master’s degree in professional computing and artificial intelligence from the Australian National University in 2019. He has been a machine learning engineer at ZTE Corporation since 2020. His research interests include machine learning, professional computing and system architecture.
Supported by:
ZTE Industry-University-Institute Cooperation Funds(HC-CN-20221123003)

摘要/Abstract

Abstract:

In recent years, microservice architecture has gained increasing popularity. However, due to the complex and dynamically changing nature of microservice systems, failure detection has become more challenging. Traditional root cause analysis methods mostly rely on a single modality of data, which is insufficient to cover all failure information. Existing multimodal methods require collecting high-quality labeled samples and often face challenges in classifying unknown failure categories. To address these challenges, this paper proposes a root cause analysis framework based on a masked graph autoencoder (GAE). The main process involves feature extraction, feature dimensionality reduction based on GAE, and online clustering combined with expert input. The method is experimentally evaluated on two public datasets and compared with two baseline methods, demonstrating significant advantages even with 16% labeled samples.

Key words: root cause analysis, multimodal data, self-supervised learning, online clustering

. [J]. ZTE Communications, 2025, 23(4): 110-119.

LI Yingke, HAN Jing, SUN Yongqian, SHI Binpeng, GONG Zican. A Root Cause Analysis Framework for Microservice Systems with Multimodal Data[J]. ZTE Communications, 2025, 23(4): 110-119.

图/表 8

参考文献 26

[1]	YI X X, ZHANG N H, LIU Y C, et al. Key technologies for intelligent computing power network [J]. ZTE technology journal, 2025, 31(2): 31–38. DOI: 10.12142/ZTETJ.202502005
[2]	WU H Q. Reflections on AI-empowered network reconstruction [J]. ZTE technology journal, 2025, 31(1): 1–3. DOI: 10.12142/ZTETJ.202501001
[3]	ZHANG S L, XIA S B, FAN W Z, et al. Failure diagnosis in microservice systems: a comprehensive survey and analysis [J]. ACM transactions on software engineering and methodology, 2025, (Just Accepted). DOI: 10.1145/3715005
[4]	JIN M X, LV A R, ZHU Y P, et al. An anomaly detection algorithm for microservice architecture based on robust principal component analysis [J]. IEEE access, 2020, 8: 226397–226408
[5]	YU G B, CHEN P F, CHEN H Y, et al. MicroRank: end-to-end latency issue localization with extended spectrum analysis in microservice environments [C]//Proceedings of the Web Conference 2021. ACM, 2021: 3087–3098. DOI: 10.1145/3442381.3449905
[6]	LIN Q W, ZHANG H Y, LOU J G, et al. Log clustering based problem identification for online service systems [C]//The 38th International Conference on Software Engineering Companion (ICSE-C). IEEE, 2016: 102–111
[7]	YUAN Y, SHI W C, LIANG B, et al. An approach to cloud execution failure diagnosis based on exception logs in OpenStack [C]//Proceedings of IEEE 12th International Conference on Cloud Computing (CLOUD). IEEE, 2019: 124–131. DOI: 10.1109/cloud.2019.00031
[8]	ZHANG X, XU Y, QIN S, et al. Onion: identifying incident-indicating logs for cloud systems [C]//The 29th ACM Joint Meeting on European Software Engineering Conference and Symposium on the Foundations of Software Engineering. ACM, 2021: 1253–1263. DOI: 10.1145/3468264.3473919
[9]	DU M, LI F F, ZHENG G N, et al. DeepLog: anomaly detection and diagnosis from system logs through deep learning [C]//Proceedings of the 2017 ACM SIGSAC Conference on Computer and Communications Security. ACM, 2017: 1285–1298. DOI: 10.1145/3133956.3134015
[10]	MA M, LIN W L, PAN D S, et al. Self-adaptive root cause diagnosis for large-scale microservice architecture [J]. IEEE transactions on services computing, 2022, 15(3): 1399–1410. DOI: 10.1109/TSC.2020.2993251
[11]	LIN J J, CHEN P F, ZHENG Z B. Microscope: pinpoint performance issues with causal graphs in micro-service environments [M]//Service-oriented computing. Cham: Springer International Publishing, 2018: 3–20. DOI: 10.1007/978-3-030-03596-9_1
[12]	MA M, XU J M, WANG Y, et al. AutoMAP: diagnose your microservice-based web applications automatically [C]//Proceedings of The Web Conference 2020. ACM, 2020: 246–258. DOI: 10.1145/3366423.3380111
[13]	ZHANG S, JIN P, LIN Z, et al. Robust root cause analysis of microservice system through multimodal data [J]. IEEE transactions on services computing, 2023:1–14
[14]	SUN Y Q, SHI B P, MAO M Y, et al. ART: a unified unsupervised framework for incident management in microservice systems [C]//Proceedings of the 39th IEEE/ACM International Conference on Automated Software Engineering. ACM, 2024: 1183–1194. DOI: 10.1145/3691620.3695495
[15]	SUN Y Q, LIN Z H, SHI B P, et al. Interpretable failure localization for microservice systems based on graph autoencoder [J]. ACM transactions on software engineering and methodology, 2025, 34(2): 52. DOI: 10.1145/3695999
[16]	LEE C, YANG T Y, CHEN Z B, et al. Eadro: an end-to-end troubleshooting framework for microservices on multi-source data [C]//The 45th International Conference on Software Engineering (ICSE). IEEE, 2023: 1750–1762. DOI: 10.1109/ICSE48619.2023.00150
[17]	ZHANG Y Y, GUAN Z X, QIAN H J, et al. CloudRCA: a root cause analysis framework for cloud computing platforms [C]//The 30th ACM International Conference on Information & Knowledge Management. ACM, 2021: 4373–4382. DOI: 10.1145/3459637.3481903
[18]	WANG C, PAN S R, LONG G D, et al. Mgae: marginalized graph autoencoder for graph clustering [C]//Conference on Information and Knowledge Management. ACM, 2017: 889–898
[19]	HOU C J, JIA T, WU Y F, et al. Diagnosing performance issues in microservices with heterogeneous data source [C]//Proceedings of IEEE Intl Conf on Parallel & Distributed Processing with Applications, Big Data & Cloud Computing, Sustainable Computing & Communications, Social Computing & Networking. IEEE, 2021: 493–500. DOI: 10.1109/ISPA-BDCloud-SocialCom-SustainCom52081.2021.00074
[20]	HOU Z Y, LIU X, CEN Y K, et al. GraphMAE: self-supervised masked graph autoencoders [C]//The 28th ACM SIGKDD Conference on Knowledge Discovery and Data Mining. ACM, 2022: 594–604. DOI: 10.1145/3534678.3539321
[21]	SIGELMAN B H, BARROSO L A, BURROWS M, et al. Dapper, a large-scale distributed systems tracing infrastructure [EB/OL]. [2024-01-26]
[22]	Opentracing. Opentracing [EB/OL]. [2024-01-26].
[23]	SIFFER A, FOUQUE P A, TERMIER A, et al. Anomaly detection in streams with extreme value theory [C]//The 23rd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. ACM, 2017: 1067–1075. DOI: 10.1145/3097983.3098144
[24]	KIPF T N, WEWELLING M. Variational graph auto-encoders [EB/OL]. (2016-11-21) [2024-01-26].
[25]	HE P J, ZHU J M, ZHENG Z B, et al. Drain: an online log parsing approach with fixed depth tree [C]//International Conference on Web Services (ICWS). IEEE, 2017: 33–40. DOI: 10.1109/ICWS.2017.13
[26]	PARK J, LEE M, CHANG H J, et al. Symmetric graph convolutional autoencoder for unsupervised graph representation learning. (2018-09-27) [2024-01-26].

Dataset	D1		D2
Dataset	Training Set	Testing Set	Training Set	Testing Set
Number of failure Cases	139	35	127	32
Number of system microservices	10		9
Number of microservice instances	20		18
Number of involved failure categories	5		5

Dataset	D1		D2
Dataset	Training Set	Testing Set	Training Set	Testing Set
Number of failure Cases	139	35	127	32
Number of system microservices	10		9
Number of microservice instances	20		18
Number of involved failure categories	5		5

Dataset	D1				D2
Approach	Precision	Recall	F1-Score	Labeled Ratio	Precision	Recall	F1-Score	Labeled Ratio
DiagFusion	0.955	0.239	0.382	0.161	0.879	0.424	0.572	0.163
CloudRCA	0.160	0.154	0.191	0.161	0.382	0.367	0.361	0.163
Our method	0.833	0.879	0.851	0.161	0.839	0.824	0.824	0.163

Dataset	D1				D2
Approach	Precision	Recall	F1-Score	Labeled Ratio	Precision	Recall	F1-Score	Labeled Ratio
DiagFusion	0.955	0.239	0.382	0.161	0.879	0.424	0.572	0.163
CloudRCA	0.160	0.154	0.191	0.161	0.382	0.367	0.361	0.163
Our method	0.833	0.879	0.851	0.161	0.839	0.824	0.824	0.163

Dataset	22 AIOps Dataset
Approach	Precision	Recall	F1-Score
OM w/o trace data	0.733	0.799	0.754
OM w/o log data	0.710	0.747	0.713
OM w/o metric data	0.256	0.506	0.340
OM w/o GAE	0.738	0.678	0.625
OM	0.833	0.879	0.851

A Root Cause Analysis Framework for Microservice Systems with Multimodal Data

RichHTML

PDF

可视化

摘要/Abstract

引用本文

使用本文

图/表 8

参考文献 26

相关文章 0

编辑推荐

Metrics