KRCA: An Efficient Root Cause Analysis System in Hyper-Scale Microservice Systems via Agentic AI

2026-07-02

AI Summary

This work addresses the inefficiency of existing root cause analysis methods in ultra-large-scale microservice systems, which stems from their high dynamism and complex dependencies. The authors propose KRCA, an end-to-end root cause analysis system that innovatively integrates structured causal priors with memory-augmented multi-agent reasoning. KRCA narrows the search space via API-level drill-down, constructs a skeleton causal graph from anomalous metrics, and leverages collaborative multi-agent inference to validate causal relationships and generate diagnostic reports. Experimental results demonstrate that KRCA achieves AC@1 scores of 0.88 for root cause service localization and 0.79 for fault type classification, representing an absolute improvement of over 31% compared to the strongest baseline. Deployed in Kuaishou’s production environment for six months, KRCA has consistently reduced average diagnosis time by 77.3%.

Abstract

Hyper-scale microservice systems have become the standard infrastructure for large-scale Internet companies. These systems consist of numerous loosely coupled microservices that evolve independently through continuous development and deployment. Such complexity makes failures unavoidable, necessitating efficient Root Cause Analysis (RCA) to help Site Reliability Engineers (SREs) quickly localize root cause services and classify failure types. However, existing RCA methods often struggle to adapt to the extreme dynamism and massive scale of these systems. In this paper, we present KRCA, an end-to-end RCA system designed for hyper-scale microservice systems. To manage the vast search space, KRCA employs a multi-stage pipeline that begins with an API-level drilldown to isolate suspicious services. It then instantiates a skeleton-based causal graph from anomalous metrics to serve as a high-recall structural prior, before utilizing a memory-augmented multi-agent framework to verify causality and generate the final failure report. By combining structured causal constraints with multi-agent reasoning, KRCA employs balances diagnostic accuracy with the efficiency requirements of real-time production use. Experimental results show that KRCA achieves AC@1 scores of 0.88 and 0.79 for root cause service localization and failure type classification, outperforming the strongest baseline by at lease 31% in absolute gains. KRCA has been deployed in Kuaishou's production environment for over six months, reducing the average diagnosis time by 77.3%.

Problem

• Root Cause Analysis
• Hyper-scale Microservice Systems
• Failure Localization
• Failure Classification
• Site Reliability Engineering

Innovation

• Root Cause Analysis
• Microservice Systems
• Causal Graph
• Multi-Agent AI
• Agentic AI