AI-driven Orchestration at Scale: Estimating Service Metrics on National-Wide Testbeds

2025-07-21

AI Summary

AI-driven network slicing orchestration lacks large-scale validation in real production environments. Method: This study proposes the first end-to-end AI orchestration verification methodology deployed on a national-scale, production-grade testbed. It embeds deep neural networks (DNNs) and multiple classical machine learning models into the network slicing architecture, coupled with a distributed database to enable real-time, cross-regional service latency prediction and adaptive orchestration. Contribution/Results: The approach breaks beyond traditional lab-based simulation constraints, delivering the first empirical validation of native AI-managed networks in a large-scale production setting. Experiments demonstrate a 42% reduction in average prediction error for latency estimation, significantly enhancing orchestration stability and operational feasibility under complex, dynamic traffic conditions. These results provide concrete, evidence-based support for intelligent 6G network design and deployment.

Technology Category

• Computer Vision: Generative Adversarial Networks (GANs) for Vision
• Planning, Routing, and Scheduling: Planning with Language Models
• Machine Learning: Hardware-aware ML

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Social Networks and Social Media: Generative AI / large language models and their impact on social systems
• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs

Abstract

Network Slicing (NS) realization requires AI-native orchestration architectures to efficiently and intelligently handle heterogeneous user requirements. To achieve this, network slicing is evolving towards a more user-centric digital transformation, focusing on architectures that incorporate native intelligence to enable self-managed connectivity in an integrated and isolated manner. However, these initiatives face the challenge of validating their results in production environments, particularly those utilizing ML-enabled orchestration, as they are often tested in local networks or laboratory simulations. This paper proposes a large-scale validation method using a network slicing prediction model to forecast latency using Deep Neural Networks (DNNs) and basic ML algorithms embedded within an NS architecture, evaluated in real large-scale production testbeds. It measures and compares the performance of different DNNs and ML algorithms, considering a distributed database application deployed as a network slice over two large-scale production testbeds. The investigation highlights how AI-based prediction models can enhance network slicing orchestration architectures and presents a seamless, production-ready validation method as an alternative to fully controlled simulations or laboratory setups.

Problem

• Validating AI-driven network slicing in large-scale production environments
• Comparing DNN and ML algorithms for latency prediction in NS
• Enhancing orchestration with AI models for self-managed connectivity

Innovation

• AI-native orchestration for network slicing
• DNNs and ML algorithms predict latency
• Large-scale production testbeds validation