Download PDFOpen PDF in browser

Joint Optimization Scheme of Multi-Service Replication and Request Offloading in Mobile Edge Computing

EasyChair Preprint 7230

15 pagesDate: December 17, 2021

Abstract

To meet the ever-increasing service quality requirements of end-users and enable delay-sensitive applications to be completed within a tolerable time, Mobile Edge Computing (MEC) offloads the request of users to the edge servers that are closer to the end equipment. However, deploying a single service replication in an appropriate edge node is difficult to deal with all requests of users for multiple services. In addition, after the service replication is deployed at the edge node, a corresponding user request offloading scheme is also required. Considering the heterogeneity of edge servers, this paper studies the joint optimization problem of multi-service replication and request offloading. Firstly, we present an edge computing architecture with multi-service replication, and define the multi-service replication and request offloading as a joint optimization problem. Secondly, a multi-service replication algorithm called Multireplicas Greedy Best (MGB) is proposed to solve the joint optimization problem. Finally, the simulation experiments are carried out. The experimental results show that the proposed algorithm can effectively reduce the overall delay compared with the random strategy, the nearest node offloading strategy, the particle swarm algorithm, and the greedy algorithm.

Keyphrases: Mobile Edge Computing, Request Offloading, Service Replication, multi-objective optimization

BibTeX entry
BibTeX does not have the right entry for preprints. This is a hack for producing the correct reference:
@booklet{EasyChair:7230,
  author    = {Chenxi Li and Guanghui Li and Shihong Hu and Chenglong Dai and Dong Li},
  title     = {Joint Optimization Scheme of Multi-Service Replication and Request Offloading in Mobile Edge Computing},
  howpublished = {EasyChair Preprint 7230},
  year      = {EasyChair, 2021}}
Download PDFOpen PDF in browser