Internet traffic load balancing using dynamic hashing with flow volume

Ju Yeon Jo, Yoohwan Kim, H. Jonathan Chao, Frank Merat

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Sending IP packets over multiple parallel links is in extensive use in today's Internet and its use is growing due to its scalability, reliability and cost-effectiveness. To maximize the efficiency of parallel links, load balancing is necessary among the links, but it may cause the problem of packet reordering. Since packet reordering impairs TCP performance, it is important to reduce the amount of reordering. Hashing offers a simple solution to keep the packet order by sending a flow over a unique link, but static hashing does not guarantee an even distribution of the traffic amount among the links, which could lead to packet loss under heavy load. Dynamic hashing offers some degree of load balancing but suffers from load fluctuations and excessive packet reordering. To overcome these shortcomings, we have enhanced the dynamic hashing algorithm to utilize the flow volume information in order to reassign only the appropriate flows. This new method, called dynamic hashing with flow volume (DHFV), eliminates unnecessary flow reassignments of small flows and achieves load balancing very quickly without load fluctuation by accurately predicting the amount of transferred load between the links. In this paper we provide the general framework of DHFV and address the challenges in implementing DHFV. We then introduce two algorithms of DHFV with different flow selection strategies and show their performances through simulation.

Original languageEnglish (US)
Title of host publicationProceedings of SPIE - The International Society for Optical Engineering
EditorsR.D. Mei, F. Huebner
Pages154-165
Number of pages12
Volume4865
DOIs
StatePublished - 2002
EventInternet Performance and Control of Network Systems III - Boston, MA, United States
Duration: Jul 30 2002Jul 31 2002

Other

OtherInternet Performance and Control of Network Systems III
CountryUnited States
CityBoston, MA
Period7/30/027/31/02

Fingerprint

traffic
Resource allocation
Internet
Cost effectiveness
Packet loss
Scalability
cost effectiveness
causes

Keywords

  • Flow analysis
  • Hashing
  • Internet
  • Load balancing
  • Multilink
  • Packet reordering

ASJC Scopus subject areas

  • Electrical and Electronic Engineering
  • Condensed Matter Physics

Cite this

Jo, J. Y., Kim, Y., Chao, H. J., & Merat, F. (2002). Internet traffic load balancing using dynamic hashing with flow volume. In R. D. Mei, & F. Huebner (Eds.), Proceedings of SPIE - The International Society for Optical Engineering (Vol. 4865, pp. 154-165) https://doi.org/10.1117/12.473386

Internet traffic load balancing using dynamic hashing with flow volume. / Jo, Ju Yeon; Kim, Yoohwan; Chao, H. Jonathan; Merat, Frank.

Proceedings of SPIE - The International Society for Optical Engineering. ed. / R.D. Mei; F. Huebner. Vol. 4865 2002. p. 154-165.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Jo, JY, Kim, Y, Chao, HJ & Merat, F 2002, Internet traffic load balancing using dynamic hashing with flow volume. in RD Mei & F Huebner (eds), Proceedings of SPIE - The International Society for Optical Engineering. vol. 4865, pp. 154-165, Internet Performance and Control of Network Systems III, Boston, MA, United States, 7/30/02. https://doi.org/10.1117/12.473386
Jo JY, Kim Y, Chao HJ, Merat F. Internet traffic load balancing using dynamic hashing with flow volume. In Mei RD, Huebner F, editors, Proceedings of SPIE - The International Society for Optical Engineering. Vol. 4865. 2002. p. 154-165 https://doi.org/10.1117/12.473386
Jo, Ju Yeon ; Kim, Yoohwan ; Chao, H. Jonathan ; Merat, Frank. / Internet traffic load balancing using dynamic hashing with flow volume. Proceedings of SPIE - The International Society for Optical Engineering. editor / R.D. Mei ; F. Huebner. Vol. 4865 2002. pp. 154-165
@inproceedings{71aa03d0ecc94237a4f2208a55bb66b5,
title = "Internet traffic load balancing using dynamic hashing with flow volume",
abstract = "Sending IP packets over multiple parallel links is in extensive use in today's Internet and its use is growing due to its scalability, reliability and cost-effectiveness. To maximize the efficiency of parallel links, load balancing is necessary among the links, but it may cause the problem of packet reordering. Since packet reordering impairs TCP performance, it is important to reduce the amount of reordering. Hashing offers a simple solution to keep the packet order by sending a flow over a unique link, but static hashing does not guarantee an even distribution of the traffic amount among the links, which could lead to packet loss under heavy load. Dynamic hashing offers some degree of load balancing but suffers from load fluctuations and excessive packet reordering. To overcome these shortcomings, we have enhanced the dynamic hashing algorithm to utilize the flow volume information in order to reassign only the appropriate flows. This new method, called dynamic hashing with flow volume (DHFV), eliminates unnecessary flow reassignments of small flows and achieves load balancing very quickly without load fluctuation by accurately predicting the amount of transferred load between the links. In this paper we provide the general framework of DHFV and address the challenges in implementing DHFV. We then introduce two algorithms of DHFV with different flow selection strategies and show their performances through simulation.",
keywords = "Flow analysis, Hashing, Internet, Load balancing, Multilink, Packet reordering",
author = "Jo, {Ju Yeon} and Yoohwan Kim and Chao, {H. Jonathan} and Frank Merat",
year = "2002",
doi = "10.1117/12.473386",
language = "English (US)",
volume = "4865",
pages = "154--165",
editor = "R.D. Mei and F. Huebner",
booktitle = "Proceedings of SPIE - The International Society for Optical Engineering",

}

TY - GEN

T1 - Internet traffic load balancing using dynamic hashing with flow volume

AU - Jo, Ju Yeon

AU - Kim, Yoohwan

AU - Chao, H. Jonathan

AU - Merat, Frank

PY - 2002

Y1 - 2002

N2 - Sending IP packets over multiple parallel links is in extensive use in today's Internet and its use is growing due to its scalability, reliability and cost-effectiveness. To maximize the efficiency of parallel links, load balancing is necessary among the links, but it may cause the problem of packet reordering. Since packet reordering impairs TCP performance, it is important to reduce the amount of reordering. Hashing offers a simple solution to keep the packet order by sending a flow over a unique link, but static hashing does not guarantee an even distribution of the traffic amount among the links, which could lead to packet loss under heavy load. Dynamic hashing offers some degree of load balancing but suffers from load fluctuations and excessive packet reordering. To overcome these shortcomings, we have enhanced the dynamic hashing algorithm to utilize the flow volume information in order to reassign only the appropriate flows. This new method, called dynamic hashing with flow volume (DHFV), eliminates unnecessary flow reassignments of small flows and achieves load balancing very quickly without load fluctuation by accurately predicting the amount of transferred load between the links. In this paper we provide the general framework of DHFV and address the challenges in implementing DHFV. We then introduce two algorithms of DHFV with different flow selection strategies and show their performances through simulation.

AB - Sending IP packets over multiple parallel links is in extensive use in today's Internet and its use is growing due to its scalability, reliability and cost-effectiveness. To maximize the efficiency of parallel links, load balancing is necessary among the links, but it may cause the problem of packet reordering. Since packet reordering impairs TCP performance, it is important to reduce the amount of reordering. Hashing offers a simple solution to keep the packet order by sending a flow over a unique link, but static hashing does not guarantee an even distribution of the traffic amount among the links, which could lead to packet loss under heavy load. Dynamic hashing offers some degree of load balancing but suffers from load fluctuations and excessive packet reordering. To overcome these shortcomings, we have enhanced the dynamic hashing algorithm to utilize the flow volume information in order to reassign only the appropriate flows. This new method, called dynamic hashing with flow volume (DHFV), eliminates unnecessary flow reassignments of small flows and achieves load balancing very quickly without load fluctuation by accurately predicting the amount of transferred load between the links. In this paper we provide the general framework of DHFV and address the challenges in implementing DHFV. We then introduce two algorithms of DHFV with different flow selection strategies and show their performances through simulation.

KW - Flow analysis

KW - Hashing

KW - Internet

KW - Load balancing

KW - Multilink

KW - Packet reordering

UR - http://www.scopus.com/inward/record.url?scp=0036422103&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0036422103&partnerID=8YFLogxK

U2 - 10.1117/12.473386

DO - 10.1117/12.473386

M3 - Conference contribution

VL - 4865

SP - 154

EP - 165

BT - Proceedings of SPIE - The International Society for Optical Engineering

A2 - Mei, R.D.

A2 - Huebner, F.

ER -