Data-driven Online Slice Admission Control and Resource Allocation for 5G and Beyond Networks

TL;DR

This work addresses online slice admission control and resource allocation in 5G+ networks where slice requests arrive sequentially and QoS-resource mappings are uncertain. It introduces OSARA, a data-driven framework that couples an online price-based admission policy with a differentiable end-to-end slice model and a gradient-based resource allocator using Lagrangian duality, providing a bound on the competitive ratio. Empirical evaluation on a real 5G testbed demonstrates significant improvements in empirical competitive ratio (up to 42%) and fast, near-optimal resource allocation compared to offline baselines, while maintaining explainability through cost-revenue analyses. The approach has practical impact for infrastructure providers seeking real-time, explainable, and provably robust slice admission and resource management in next-generation networks.

Abstract

Virtualization in 5G and beyond networks allows the creation of virtual networks, or network slices, tailored to meet the requirements of various applications. However, this flexibility introduces several challenges for infrastructure providers (InPs) in slice admission control (AC) and resource allocation. To maximize revenue, InPs must decide in real-time whether to admit new slice requests (SRs) given slices' revenues, limited infrastructure resources, unknown relationship between resource allocation and Quality of Service (QoS), and the unpredictability of future SRs. To address these challenges, this paper introduces a novel data-driven framework for 5G slice admission control that offers a guaranteed upper bound on the competitive ratio, i.e., the ratio between the revenue obtained by an oracle solution and that of the online solution. The proposed framework leverages a pricing function to dynamically estimate resources' pseudo-prices that reflect resource scarcity. Such prices are further coupled with a resource allocation algorithm, which leverages a machine-learned slice model and employs a primal-dual algorithm to determine the minimum-cost resource allocation. The resource cost is then compared with the offered revenue to admit or reject a SR. To demonstrate the efficacy of our framework, we train the data-driven slice model using real traces collected from our 5G testbed. Our results show that our novel approach achieves up to 42% improvement in the empirical competitive ratio, i.e., ratio between the optimal and the online solution, compared to other benchmark algorithms.

Figures (9)

• Figure 1: Overview of OPA for OSARA
• Figure 2: Slice and VNF model architectures
• Figure 3: Overview of our 5G testbed
• Figure 4: VNF model loss curves
• Figure 5: Slice model regression plot

Theorems & Definitions (2)

• Theorem 1
• proof : Proof of Theorem \ref{['thm:osac']}