Enabling Autonomic Microservice Management through Self-Learning Agents

TL;DR

ServiceOdyssey addresses the challenge of autonomic microservice management by enabling self-learning agents to acquire service-specific knowledge without prior configurations. It introduces a three-module management stack—Curriculum Builder, Execution Planner, and Knowledge Curator—built atop a data layer to support curriculum-driven exploration, executable plan refinement, and a growing skill library. The approach is demonstrated via a Sock Shop prototype, showing progressive knowledge acquisition, low-cost iterations, and reduced reliance on human input. This work highlights a scalable pathway toward autonomous microservice governance in complex distributed environments, with future work on parallel exploration and memory-efficient skill sharing.

Abstract

The increasing complexity of modern software systems necessitates robust autonomic self-management capabilities. While Large Language Models (LLMs) demonstrate potential in this domain, they often face challenges in adapting their general knowledge to specific service contexts. To address this limitation, we propose ServiceOdyssey, a self-learning agent system that autonomously manages microservices without requiring prior knowledge of service-specific configurations. By leveraging curriculum learning principles and iterative exploration, ServiceOdyssey progressively develops a deep understanding of operational environments, reducing dependence on human input or static documentation. A prototype built with the Sock Shop microservice demonstrates the potential of this approach for autonomic microservice management.

Figures (4)

• Figure 1: An overview of the ServiceOdyssey system.
• Figure 2: Agentic microservice system for Sock Shop.
• Figure 3: Knowledge curation during self-learning.
• Figure 4: Performance evaluation in Trial 1 after each exploration round. The Y-axis lists evaluation tasks; each cell’s “x/y” indicates the number of successful trials out of the total.