Eadro: An End-to-End Troubleshooting Framework for Microservices on Multi-source Data
The complexity and dynamism of microservices pose significant challenges to system reliability, and thereby, automated troubleshooting is crucial. However, two significant issues rest in existing approaches: (1) Microservices generate traces, system logs, and key performance indicators (KPIs), but existing approaches usually consider traces only, failing to understand the system fully as traces cannot depict all anomalies; (2) Troubleshooting microservices generally contains two main phases, i.e., anomaly detection and root cause localization. Existing studies regard these two phases as independent, ignoring their close correlation. Even worse, inaccurate detection results can deeply affect localization effectiveness. To overcome these limitations, we propose Eadro, the first end-to-end framework to integrate anomaly detection and root cause localization based on multi-source data for troubleshooting large-scale microservices. The key insights of Eadro are the anomaly manifestations on different data sources and the close connection between detection and localization. Thus, Eadro models intra-service behaviors and inter-service dependencies from traces, logs, and KPIs, all the while leveraging the shared knowledge of the two phases via multi-task learning. Experiments on two widely-used benchmark microservices demonstrate that Eadro outperforms state-of-the-art approaches by a large margin, and show the effectiveness of integrating multi-source data. We also release our code and data to facilitate future research.
Fri 19 MayDisplayed time zone: Hobart change
11:00 - 12:30 | Runtime analysis and self-adaptationTechnical Track / NIER - New Ideas and Emerging Results / SEIP - Software Engineering in Practice / Journal-First Papers at Level G - Plenary Room 1 Chair(s): Domenico Bianculli University of Luxembourg | ||
11:00 15mTalk | Heterogeneous Anomaly Detection for Software Systems via Semi-supervised Cross-modal Attention Technical Track Cheryl Lee The Chinese University of Hong Kong, Tianyi Yang The Chinese University of Hong Kong, Zhuangbin Chen Chinese University of Hong Kong, China, Yuxin Su Sun Yat-sen University, Yongqiang Yang Huawei Technologies, Michael Lyu The Chinese University of Hong Kong Pre-print | ||
11:15 15mTalk | Recommending Root-Cause and Mitigation Steps for Cloud Incidents using Large Language Models Technical Track Toufique Ahmed University of California at Davis, Supriyo Ghosh Microsoft, Chetan Bansal Microsoft Research, Thomas Zimmermann Microsoft Research, Xuchao Zhang Microsoft, Saravanakumar Rajmohan Microsoft 365 Pre-print | ||
11:30 15mTalk | Eadro: An End-to-End Troubleshooting Framework for Microservices on Multi-source Data Technical Track Cheryl Lee The Chinese University of Hong Kong, Tianyi Yang The Chinese University of Hong Kong, Zhuangbin Chen Chinese University of Hong Kong, China, Yuxin Su Sun Yat-sen University, Michael Lyu The Chinese University of Hong Kong Pre-print | ||
11:45 15mTalk | LogReducer: Identify and Reduce Log Hotspots in Kernel on the Fly Technical Track Guangba Yu Sun Yat-Sen University, Pengfei Chen Sun Yat-Sen University, Pairui Li Tencent Inc., Tianjun Weng Tencent Inc., Haibing Zheng Tencent, Yuetang Deng Tencent, Zibin Zheng School of Software Engineering, Sun Yat-sen University Pre-print | ||
12:00 15mTalk | TraceArk: Towards Actionable Performance Anomaly Alerting for Online Service Systems SEIP - Software Engineering in Practice Zhengran Zeng Southern University of Science and Technology, Yuqun Zhang Southern University of Science and Technology, Yong Xu Microsoft Research, Minghua Ma Microsoft Research, Bo Qiao Microsoft Research, Wentao Zou , Qingjun Chen , Meng Zhang , Xu Zhang Microsoft Research, Hongyu Zhang The University of Newcastle, Xuedong Gao , Hao Fan , Saravan Rajmohan Microsoft 365, Qingwei Lin Microsoft Research, Dongmei Zhang Microsoft Research | ||
12:15 7mTalk | ActivFORMS: A Formally-Founded Model-Based Approach to Engineer Self-Adaptive Systems Journal-First Papers | ||
12:22 7mTalk | Auto-Logging: AI-centred Logging Instrumentation NIER - New Ideas and Emerging Results Pre-print |