一句话：提出 Chimera，通过异常检测与根因定位在数据、特征与结果层的双向交互，实现端到端日志故障诊断。
基于日志的故障诊断对保障系统可用性至关重要，但现有方法通常将异常检测与根因定位作为独立任务构建，二者在数据形式与诊断目标上彼此割裂，导致诊断偏差在系统中累积、部署依赖昂贵的监控数据、且忽视任务间协作关系。为此，作者提出端到端日志故障诊断方法 Chimera，其核心思想是通过异常检测与根因定位之间的双向交互与知识迁移实现端到端诊断。Chimera 基于交互式多任务学习，在数据层、特征层和诊断结果层精心设计交互策略，使两个子任务在统一的端到端框架内相互促进。在两个公开数据集与一个工业数据集上的评估表明，Chimera 在异常检测和根因定位上均优于现有方法，分别取得 2.92%~5.00% 与 19.01%~37.09% 的提升，并已成功部署于生产环境的工业云平台。
Log-based fault diagnosis is essential for maintaining software system availability. However, existing fault diagnosis methods are built using a task-independent manner, which fails to bridge the gap between anomaly detection and root cause localization in terms of data form and diagnostic objectives, resulting in three major issues: 1) Diagnostic bias accumulates in the system; 2) System deployment relies on expensive monitoring data; 3) The collaborative relationship between diagnostic tasks is overlooked. Facing this problems, we propose a novel end-to-end log-based fault diagnosis method, Chimera, whose key idea is to achieve end-to-end fault diagnosis through bidirectional interaction and knowledge transfer between anomaly detection and root cause localization. Chimera is based on interactive multitask learning, carefully designing interaction strategies between anomaly detection and root cause localization at the data, feature, and diagnostic result levels, thereby achieving both sub-tasks interactively within a unified end-to-end framework. Evaluation on two public datasets and one industrial dataset shows that Chimera outperforms existing methods in both anomaly detection and root cause localization, achieving improvements of over 2.92%~5.00% and 19.01% ~ 37.09%, respectively. It has been successfully deployed in production, serving an industrial cloud platform.
DOI 原文 ·
@inproceedings{paperbot9245,
title = {United We Stand: Towards End-to-End Log-based Fault Diagnosis via Interactive Multi-Task Learning},
author = {Minghua He and Chiming Duan and Pei Xiao and Tong Jia and Siyu Yu and Lingzhe Zhang and Weijie Hong and Jin Han and Yifan Wu and Ying Li and Gang Huang},
booktitle = {ASE},
year = {2025},
doi = {10.1109/ase63991.2025.00061}
}