Microservice logs analysis employing AI: A systematic literature review

Abstract

Background: Microservice architectures generate massive volumes of fragmented log data. While complicating unified monitoring and diagnosis, traditional approaches are typically overwhelmed, leading to delayed incident detection and costly system failures. Artificial Intelligence (AI) techniques, particularly machine learning and large language models, have emerged as promising solutions for automating log analysis and addressing these operational challenges. Objective: The objective of this study is to systematically review and synthesize existing research on AI techniques for microservice log analysis, evaluating their enterprise deployment readiness and identifying current capabilities, limitations, and priority areas for future investigation. Method: We conducted a systematic literature review of 2208 papers from peer-reviewed academic literature, published between 2018 and 2025. Through a rigorous filtering process, we identified 82 primary studies that thoroughly examine the use of AI for microservice log analysis across tools, techniques, datasets, and challenges. Results: While AI techniques are frequently applied to anomaly detection and root cause analysis, this review reveals a critical disconnect between research advances and industry needs. 65 studies rely on synthetic or private datasets that poorly reflect production complexities. Approximately two-thirds (65.85%) of papers adopt standardized evaluation benchmarks, with 59.76% using the standard Precision-Recall-F1 classification metrics. Hence, unresolved scalability, generalizability, and data accessibility remain the key challenges that prevent widespread adoption of AI-based microservice log analysis in enterprise environments. Conclusion: AI has strong potential for advancing log analysis in microservices. Future research should prioritize open datasets, robust benchmarking, and the use of these AI methods for deeper contextual understanding such as understanding why anomalies propagate across services and how they may impact business operations.