Improved Routing in Networks through Load Prediction Strategy

Ron Posti, Michael Segal

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

Abstract

Communication protocols are mostly based on shortest-path and per-flow static hashing which may cause under utilization, congestion and misusage of links. Thanks to the advancement of Software Defined Networking (SDN), computing and storage technologies, today's big data analytics, machine learning and artificial intelligence (AI) technologies give network operators an unprecedented opportunity to gain network insights and move towards network autonomy. This work presents a proof for the efficiency of prediction algorithms in addition to their load prediction accuracy lower bound required to help network operators to improve the overall network performance. The results show us that the load prediction accuracy lower bound does not necessarily depend on the network topologies' parameters.

Original languageEnglish
Title of host publication2021 17th International Conference on the Design of Reliable Communication Networks, DRCN 2021
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9780738123288
DOIs
StatePublished - 19 Apr 2021
Event17th International Conference on the Design of Reliable Communication Networks, DRCN 2021 - Milano, Italy
Duration: 19 Apr 202122 Apr 2021

Publication series

Name2021 17th International Conference on the Design of Reliable Communication Networks, DRCN 2021

Conference

Conference17th International Conference on the Design of Reliable Communication Networks, DRCN 2021
Country/TerritoryItaly
CityMilano
Period19/04/2122/04/21

Keywords

  • ECMP
  • SDN
  • load balancing
  • load prediction

ASJC Scopus subject areas

  • Computer Networks and Communications
  • Hardware and Architecture
  • Safety, Risk, Reliability and Quality

Fingerprint

Dive into the research topics of 'Improved Routing in Networks through Load Prediction Strategy'. Together they form a unique fingerprint.

Cite this