REACH: Reinforcement Learning for Adaptive Microservice Rescheduling in the Cloud-Edge Continuum

2025-10-08

AI Summary

To address the challenge of simultaneously achieving low latency and stability in microservice deployment under heterogeneous and dynamic resource conditions across the cloud–edge continuum, this paper proposes an adaptive microservice rescheduling algorithm based on reinforcement learning. The method pioneers the application of deep reinforcement learning to cloud–edge collaborative scenarios, enabling autonomous decision-making—guided by real-time resource-state feedback—on service migration and redeployment to respond online to workload fluctuations and performance variations. Evaluated on a realistic distributed testbed, the approach reduces end-to-end average latency by 7.9%, 10.0%, and 8.0% for three benchmark microservice applications, respectively, while significantly suppressing latency jitter. This work establishes a scalable, adaptive paradigm for service orchestration in dynamic edge environments.

Technology Category

• Machine Learning: Learning on the Edge & Model Compression
• Planning, Routing, and Scheduling: Planning with Language Models
• Multiagent Systems: Multiagent Learning

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Cloud, edge and content delivery systems for the Web
• Responsible Web: Human-perceived consequences of algorithmic deployment on the web
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs

Abstract

Cloud computing, despite its advantages in scalability, may not always fully satisfy the low-latency demands of emerging latency-sensitive pervasive applications. The cloud-edge continuum addresses this by integrating the responsiveness of edge resources with cloud scalability. Microservice Architecture (MSA) characterized by modular, loosely coupled services, aligns effectively with this continuum. However, the heterogeneous and dynamic computing resource poses significant challenges to the optimal placement of microservices. We propose REACH, a novel rescheduling algorithm that dynamically adapts microservice placement in real time using reinforcement learning to react to fluctuating resource availability, and performance variations across distributed infrastructures. Extensive experiments on a real-world testbed demonstrate that REACH reduces average end-to-end latency by 7.9%, 10%, and 8% across three benchmark MSA applications, while effectively mitigating latency fluctuations and spikes.

Problem

• Optimizing microservice placement in heterogeneous cloud-edge environments
• Addressing latency fluctuations in dynamic distributed infrastructures
• Adapting to changing resource availability for latency-sensitive applications

Innovation

• Reinforcement learning adapts microservice placement dynamically
• Real-time rescheduling algorithm for cloud-edge continuum
• Mitigates latency fluctuations across distributed infrastructures