Year: 2025
Type: article
Abstract: The exponential growth of artificial intelligence applications has necessitated significant advancements in AI infrastructure to manage the computational, storage, and energy demands of large-scale machine learning models.This paper introduces a unified framework for optimizing resource allocation across model training and deployment phases, leveraging dynamic resource scaling, energy-efficient hardware configurations, and containerization techniques.Experimental results demonstrate up to 45% reduction in infrastructure costs and 32% improvement in energy efficiency while maintaining model performance.Case studies in e-commerce and healthcare illustrate the framework's adaptability, providing a robust pathway for sustainable AI system deployment.
Language: English
Subfield: Information Systems
Field: Computer Science
Domain: Physical Sciences
Open Access status: hybrid
OpenAlex ID: W4407409749