首页 > AI前沿 > STAR: Adaptive Spatial-Temporal Normalization for Unified Microservice Incident Management

STAR: Adaptive Spatial-Temporal Normalization for Unified Microservice Incident Management

arXiv机器学习 2026-09-13 03:08 7 阅读 查看原文

Automated incident management in large-scale microservice systems relies on learning robust representations from multimodal observability data, including metrics, logs, and traces.

Although recent self-supervised frameworks enable unified modeling for anomaly detection (AD), failure triage (FT), and root cause localization (RCL), they often struggle with non-stationary temporal dynamics and heterogeneous service dependency structures.

In this paper

We propose STAR, a Spatial-Temporal Adaptive Representation learning framework that explicitly addresses these challenges through adaptive normalizations.

STAR introduces two tightly coupled mechanisms:

  • Temporal Adaptive Normalization (TAN), which dynamically normalizes multivariate time series using multi-scale temporal context,
  • Spatial Adaptive Normalization (SAN), which performs structure-aware normalization over service dependency graphs.

Unlike prior methods that treat normalization as static or task-agnostic, STAR formulates it as a learnable, context-conditioned transformation aligned with the intrinsic properties of microservice systems.

The resulting adaptive representations are integrated into a unified self-supervised framework, enabling end-to-end unsupervised support for AD, FT, and RCL tasks.

Experiments

Extensive experiments on two real-world microservice benchmarks demonstrate that STAR consistently outperforms all state-of-the-art baselines, yielding significant and stable improvements across all three tasks.

Our results highlight adaptive normalization as a principled and effective mechanism for robust multimodal representation learning in complex software systems.