Automatic failure diagnosis is crucial for large microservice systems. Currently, most failure diagnosis methods rely solely on single-modal data (i.e., using either metrics, logs, or traces). In this study, we conduct an empirical study using real-world failure cases to show that combining these sources of data (multimodal data) leads to a more accurate diagnosis. However, effectively representing these data and addressing imbalanced failures remain challenging. To tackle these issues, we propose DiagFusion, a robust failure diagnosis approach that uses multimodal data. It leverages embedding techniques and data augmentation to represent the multimodal data of service instances, combines deployment data and traces to build a dependency graph, and uses a graph neural network to localize the root cause instance and determine the failure type. Our evaluations using real-world datasets show that DiagFusion outperforms existing methods in terms of root cause instance localization (improving by 20.9% to 368%) and failure type determination (improving by 11.0% to 169%).
@article{arxiv.2302.10512,
title = {Robust Failure Diagnosis of Microservice System through Multimodal Data},
author = {Shenglin Zhang and Pengxiang Jin and Zihan Lin and Yongqian Sun and Bicheng Zhang and Sibo Xia and Zhengdan Li and Zhenyu Zhong and Minghua Ma and Wa Jin and Dai Zhang and Zhenyu Zhu and Dan Pei},
journal= {arXiv preprint arXiv:2302.10512},
year = {2023}
}