Abstract
Failure diagnosis in modern software systems requires iterative evidence acquisition and hypothesis reasoning guided by operational experience. Existing LLM-based methods improve diagnosis through agentic reasoning or knowledge augmentation, but they often lack a mechanism to coordinate the evolving diagnostic state with operational experience during iterative diagnosis.
We propose OpsMem, a dual-memory framework that maintains a short-term memory for the current diagnostic state and a long-term memory for reusable operational experience. OpsMem uses cross-memory resonance to activate state-relevant long-term memory, conditions multi-agent diagnosis on the short-term and activated long-term memories, and consolidates reusable experience from solved incidents back into long-term memory.
Experiments on a real-world Huawei microservice failure diagnosis dataset show that OpsMem outperforms representative agentic-reasoning and knowledge-augmented baselines, improving Match and Relevance by up to 46.88% and 18.39% over the strongest baseline, respectively.
Blogger's Review: OpsMem effectively integrates short-term and long-term experiences through its dual-memory mechanism, offering a more flexible and efficient approach to failure diagnosis. The innovative cross-memory resonance significantly enhances diagnostic accuracy, marking an important advancement in modern software fault handling. Its performance in practical applications also provides a solid reference for future research.