Toward Energy and Location-Aware Resource Allocation in Next Generation Networks

TL;DR

This paper presents a novel market-based approach for resource allocation in next-generation networks, balancing energy constraints, fairness, and utility for multiple service providers across distributed edge and cloud resources.

Contribution

It introduces a Fisher market model for joint communication and compute resource allocation under energy constraints, ensuring high utility and fairness with low complexity.

Findings

Achieves high utility while respecting energy constraints.

Promotes fairness among service providers.

Demonstrates trade-offs between utility and energy consumption.

Abstract

Wireless networks are evolving from radio resource providers to complex systems that also involve computing, with the latter being distributed across edge and cloud facilities. Also, their optimization is shifting more and more from a performance to a value-oriented paradigm. The two aspects shall be balanced continuously, to maximize the utilities of Services Providers (SPs), users quality of experience and fairness, while meeting global constraints in terms of energy consumption and carbon footprint among others, with all these heterogeneous resources contributing. In this paper, we tackle the problem of communication and compute resource allocation under energy constraints, with multiple SPs competing to get their preferred resource bundle by spending a a fictitious currency budget. By modeling the network as a Fisher market, we propose a low complexity solution able to achieve high utilities and guarantee energy constraints, while also promoting fairness among SPs, as compared to a social optimal solution. The market equilibrium is proved mathematically, and numerical results show the multi-dimensional trade-off between utility and energy at different locations, with communication and computation-intensive services.