New system manages 5g edge resources using language and auctions
Conflict-Aware Autonomous Resource Orchestration for 5G Edge Slicing Using Multi-Agent Systems and LLMs
Networking and Internet ArchitectureDistributed, Parallel, and Cluster ComputingMultiagent Systems
Summary
Managing resources in 5G edge networks is tricky because the network needs to understand what operators want and quickly adjust limited computing and bandwidth resources. The paper presents a two-level system: one part uses a large language model (LLM) to turn operator instructions into economic policies, and another part uses an auction system among agents to fairly allocate resources based on these policies. This combined approach was tested on a real network cluster and showed it can adapt to changing demands and priorities while ensuring key guarantees.
What this means in practice
- •For telecom network operators: Automatically convert natural language service priorities into resource policies and allocate bandwidth dynamically for 5G edge network slices.
- •For cloud infrastructure teams: Implement coordinated resource control in Kubernetes clusters that host 5G user plane functions using automated bidding and CPU/bandwidth limit adjustments.
Authors
Nathan R. Prediger, Rodrigo Moreira, Juliano Araujo Wickboldt, Tereza C. Carvalho, Pedro Oliveira, Paulo Novais, Flávio de Oliveira Silva
Abstract
5G edge network slicing must reconcile two coupled requirements: understanding high-level operator intent and reallocating scarce computing and bandwidth resources under dynamic contention. Centralized optimization can face computational and responsiveness challenges under rapidly changing edge contention, and Multi-Agent System (MAS)-based auctions can support resource arbitration but still depend on operator-defined economic parameters and priorities. Embedding a Large Language Model (LLM) directly in a recurring control loop, by contrast, may introduce non-negligible inference latency and non-deterministic behavior. This paper proposes a two-timescale architecture that addresses this gap: an LLM-based Strategist translates natural-language network intents into economic policy parameters (budget, income, and bidding-aggressiveness) at the macro scale, while a deterministic, Vickrey-inspired second-price mechanism run by SPADE-based slice agents resolves recurring resource contention at the operational scale, with the resulting allocations enforced by patching Kubernetes CPU limits and Calico CNI bandwidth annotations directly on User Plane Function pods. We evaluate the architecture end-to-end on a real Kubernetes/Open5GS edge cluster across two complementary scenarios, each repeated 30 times: a Smart-City scenario, in which natural-language public-safety intents propagate into measurable changes in slice bids and bandwidth allocations within a few auction rounds, and a Hierarchical-slicing scenario, in which the mechanism provides priority-dependent resource and QoE differentiation, adapts to abrupt demand changes, and preserves configured minimum guarantees while maintaining near-complete allocation of the shared bandwidth pool.