Control plane latency reduction for service chaining in mobile edge computing system

Chi Hsiang Hung, Yao Chou Hsieh, Li Chun Wang

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

2 Scopus citations

Abstract

Software-Defined Networking and Network Function Virtualization technologies are utilized to construct service chains cheaply and elastically. In Mobile Edge Computing platform, cloud servers can be deployed near to base stations. Migrating service functions of service chains to edge can reduce network latency or save unnecessary bandwidth consumption between edge and cloud datacenter. However, the control plane latency of SDN/NFV Mobile Edge Computing Platform is subject to the scalability of flow table in SDN switches due to flow table overflow problem. In this paper, we proposed Hash-based Group Management scheme to reduce the number of maintained flow entries by assigning users into groups and design a hash-based structure to efficiently maintain the group information. In our simulation results, the proposed scheme can reduce the number of maintain flow entries and table overflow ratio to improve the control plane latency.

Original languageEnglish
Title of host publication2017 13th International Conference on Network and Service Management, CNSM 2017
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages1-5
Number of pages5
ISBN (Electronic)9783901882982
DOIs
StatePublished - 1 Jul 2017
Event13th International Conference on Network and Service Management, CNSM 2017 - Tokyo, Japan
Duration: 26 Nov 201730 Nov 2017

Publication series

Name2017 13th International Conference on Network and Service Management, CNSM 2017
Volume2018-January

Conference

Conference13th International Conference on Network and Service Management, CNSM 2017
CountryJapan
CityTokyo
Period26/11/1730/11/17

Keywords

  • Mobile Edge Computing
  • NFV
  • SDN
  • Service Chaining

Fingerprint Dive into the research topics of 'Control plane latency reduction for service chaining in mobile edge computing system'. Together they form a unique fingerprint.

Cite this