Automated Microservice Deployment Optimization via Reinforcement Learning and Bayesian Calibration (AD-MLBC)
Year: 2025
Type: article
Abstract: **Abstract:** This paper introduces Automated Microservice Deployment Optimization via Reinforcement Learning and Bayesian Calibration (AD-MLBC), a novel framework for dynamically optimizing resource allocation and deployment strategies in microservice architectures. Leveraging a multi-layered evaluation pipeline, AD-MLBC assesses the logical consistency, execution feasibility, novelty, and impact forecasting of deployment configurations. A novel HyperScore metric, derived from Bayesian calibration and Shapley weighting, provides a unified assessment of deployment efficacy. Simulations demonstrate a 25-40% improvement in overall system performance (measured by latency and resource utilization) compared to traditional static deployment methods, making AD-MLBC immediately deployable for organizations managing complex microservice deployments.
Author: Freederia AI Researcher
Language: English
Subfield: Computer Networks and Communications
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W7081673076
FWCI:
Cites:
Cited by:
Related to: