Abstract
The Multi-access Edge Computing (MEC) paradigm promises to enhance network flexibility and scalability through resource virtualization. MEC allows telecom operators to fulfill the stringent and heterogeneous requirements of 5G applications via service deployment at the edge of the mobile network. However, current solutions to support MEC struggle to provide resource elasticity since MEC infrastructures have limited resources. The coexistence of many heterogeneous services on the distributed MEC infrastructure makes the resource scarcity problem even more challenging than it already is in traditional networks. Services need distinct resource provisioning patterns due to their diverse requirements, and we may not assume an extensive MEC infrastructure that can accommodate an arbitrary number of services. To address these aspects, we present REACT: a MEC-suppoRted sElf-adaptive elAstiCiTy mechanism that leverages resource provisioning among different services running on a shared MEC environment. REACT adopts an adaptive and solidarity-based strategy to redistribute resources from over-provisioned services to under-provisioned services in MEC environments. REACT is an alternative strategy to avoid service migration due to resource scarcity. Real testbed results show that REACT outperforms Kubernetes’ elasticity strategy by accomplishing up to 18.88% more elasticity events, reducing service outages by up to 95.1%, reducing elasticity attempts by up to 95.36%, and reducing over-provisioned resources by up to 33.88%, 38.41%, and 73% for CPU cycles, RAM and bandwidth resources, respectively. Finally, REACT reduces response time by up to 15.5%.
Highlights
The realization of the 5G architecture is guided by novel technologies and new trends in user demands for modern applications, such as tactile Internet, autonomous vehicles, immersive media services, eHealth, etc [1]
Our research focuses on proposing a heuristic elasticity solution tailored to Multi-access Edge Computing (MEC) systems, capable of overcoming resource scarcity and resource over-provisioning in these systems
REACT distinguishes itself from reactive elasticity solutions in three ways: (i) optimal auto-scaling of both network-level and compute-level virtual resources at network edges under resource scarcity conditions; (ii) efficient resource allocation of over-provisioned resources from a set of donor services to scale-up demanding recipient services; and (iii) self-adaptive auto-scaling, which reduces the elasticity attempt window during the scarcity of MEC resources
Summary
The realization of the 5G architecture (including 5G beyond approaches, like 6G or Networking 2030) is guided by novel technologies and new trends in user demands for modern applications, such as tactile Internet, autonomous vehicles, immersive media services, eHealth, etc [1]. It is essential to develop elasticity strategies adapted to MEC since edge servers may run out of resources as service providers offer more resources for applications as consumer demand increases [18]. We propose the MEC-suppoRted sElf-adaptive elAstiCiTy (REACT), a new auto-scaling strategy that addresses the previously described weaknesses of reactive approaches employing a solidarity-based elasticity algorithm. REACT distinguishes itself from reactive elasticity solutions in three ways: (i) optimal auto-scaling of both network-level and compute-level virtual resources at network edges under resource scarcity conditions; (ii) efficient resource allocation of over-provisioned resources from a set of donor services to scale-up demanding recipient services; and (iii) self-adaptive auto-scaling, which reduces the elasticity attempt window during the scarcity of MEC resources. REACT could guarantee that latency-sensitive services would obtain resource reservations in MEC servers during these service migrations, classifying these services as priority services, i.e., recipient services, and the other services implemented in the MEC infrastructure as donor services
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
