LPB: A Novel Load Balance Algorithm for OPSquare DCN under Real Application Traffics

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

Abstract

Network virtualization in today's data center (DC) creates new heterogeneous traffic patterns different from what so far have been reported in literatures. To properly evaluate the DC network performance, a virtualized DC traffic model is developed from the real application traffics. In this paper, we propose a novel lowest path buffer (LPB) algorithm and evaluate the network performance of LPB in OPSquare DC under the developed traffic model. Compared with Round- Robin and Localflow, LPB could achieve 23.7% and 32.1% less latency, respectively. Besides, LPB provides lower packet loss with respect to Round-Robin, Drill and Localflow.

Original languageEnglish
Title of host publication2020 24th International Conference on Optical Network Design and Modeling, ONDM 2020
PublisherInstitute of Electrical and Electronics Engineers
ISBN (Electronic)9783903176218
DOIs
Publication statusPublished - May 2020
Event24th International Conference on Optical Network Design and Modeling, ONDM 2020 - Castelldefels, Barcelona, Spain
Duration: 18 May 202021 May 2020

Conference

Conference24th International Conference on Optical Network Design and Modeling, ONDM 2020
CountrySpain
CityCastelldefels, Barcelona
Period18/05/2021/05/20

Keywords

  • Data center network
  • Load balancing
  • Traffic modeling

Fingerprint Dive into the research topics of 'LPB: A Novel Load Balance Algorithm for OPSquare DCN under Real Application Traffics'. Together they form a unique fingerprint.

  • Cite this

    Yan, F., Guo, X., Pan, B., Xue, X., Zhang, S., & Calabretta, N. (2020). LPB: A Novel Load Balance Algorithm for OPSquare DCN under Real Application Traffics. In 2020 24th International Conference on Optical Network Design and Modeling, ONDM 2020 [9133006] Institute of Electrical and Electronics Engineers. https://doi.org/10.23919/ONDM48393.2020.9133006