Two-criteria Pareto frontier for virtual network allocation on Edge-as-a-Service networks

Abstract

Nowadays, the Internet is part of our lives, being an essential way of communication. However, it suffers from limitations that prevent the Internet to guarantee Quality of Service (QoS) for the services to the clients. Recently, the scientific community presented the Edge as a Service (EaaS) paradigm as a possible approach to improve the Internet access service. EaaS is based on the deployment of Virtual Networks (VNs) over Software Defined Networks to ease the management of Internet Service Providers (ISPs) through the supporting of Service Level Agreements (SLAs) and the allocation schemes that consider the network resources. Furthermore, the ISPs aim to maximize their profit by establishing as many SLAs as possible and decrease the energy consumption to deliver their services, where reliability and available bandwidth are key requirements. Thus, the EaaS needs an allocation algorithm to allow the deployment of VNs considering all these points. Within this context, this paper presents the PAreto based on Bandwidth and ENergy (PABEN) algorithm to deploy the VNs based on a Pareto frontier (using bandwidth availability and energy efficiency information), as well as considering a minimum reliability. The results suggest that the proposed algorithm can deploy reliable VNs, while improving bandwidth utilization and energy consumption.