Trading link utilization for queueing delays

An adaptive approach

Rade Stanojevic, Robert Shorten

Research output: Contribution to journalArticle

6 Citations (Scopus)

Abstract

Understanding the relationship between queueing delays and link utilization for general traffic conditions is an important open problem in networking research. Difficulties in understanding this relationship stem from the fact that it depends on the complex nature of arriving traffic and the problems associated with modelling such traffic. Existing AQM schemes achieve a "low delay" and "high utilization" by responding early to congestion without considering the exact relationship between delay and utilization. However, in the context of exploiting the delay/utilization tradeoff, the optimal choice of a queueing scheme's control parameter depends on the cost associated with the relative importance of queueing delay and utilization. The optimal choice of control parameter is the one that maximizes a benefit that can be defined as the difference between utilization and cost associated with queuing delay. We present two practical algorithms, Optimal Drop-Tail (ODT) and Optimal BLUE (OB), that are designed with a common performance goal: namely, maximizing this benefit. Their novelty lies in fact that they maximize the benefit in an online manner, without requiring knowledge of the traffic conditions, specific delay-utilization models, nor do they require complex parameter estimation. Packet level ns2 simulations are given to demonstrate the efficacy of the proposed algorithms and the framework in which they are designed.

Original languageEnglish
Pages (from-to)1108-1121
Number of pages14
JournalComputer Communications
Volume33
Issue number9
DOIs
Publication statusPublished - 1 Jun 2010
Externally publishedYes

Fingerprint

Parameter estimation
Costs

Keywords

  • Buffer management
  • Delay/utilization tradeoff
  • Internet routers
  • Network utility

ASJC Scopus subject areas

  • Computer Networks and Communications

Cite this

Trading link utilization for queueing delays : An adaptive approach. / Stanojevic, Rade; Shorten, Robert.

In: Computer Communications, Vol. 33, No. 9, 01.06.2010, p. 1108-1121.

Research output: Contribution to journalArticle

@article{59596b7c20434008abe7f49f37451609,
title = "Trading link utilization for queueing delays: An adaptive approach",
abstract = "Understanding the relationship between queueing delays and link utilization for general traffic conditions is an important open problem in networking research. Difficulties in understanding this relationship stem from the fact that it depends on the complex nature of arriving traffic and the problems associated with modelling such traffic. Existing AQM schemes achieve a {"}low delay{"} and {"}high utilization{"} by responding early to congestion without considering the exact relationship between delay and utilization. However, in the context of exploiting the delay/utilization tradeoff, the optimal choice of a queueing scheme's control parameter depends on the cost associated with the relative importance of queueing delay and utilization. The optimal choice of control parameter is the one that maximizes a benefit that can be defined as the difference between utilization and cost associated with queuing delay. We present two practical algorithms, Optimal Drop-Tail (ODT) and Optimal BLUE (OB), that are designed with a common performance goal: namely, maximizing this benefit. Their novelty lies in fact that they maximize the benefit in an online manner, without requiring knowledge of the traffic conditions, specific delay-utilization models, nor do they require complex parameter estimation. Packet level ns2 simulations are given to demonstrate the efficacy of the proposed algorithms and the framework in which they are designed.",
keywords = "Buffer management, Delay/utilization tradeoff, Internet routers, Network utility",
author = "Rade Stanojevic and Robert Shorten",
year = "2010",
month = "6",
day = "1",
doi = "10.1016/j.comcom.2010.02.014",
language = "English",
volume = "33",
pages = "1108--1121",
journal = "Computer Communications",
issn = "0140-3664",
publisher = "Elsevier",
number = "9",

}

TY - JOUR

T1 - Trading link utilization for queueing delays

T2 - An adaptive approach

AU - Stanojevic, Rade

AU - Shorten, Robert

PY - 2010/6/1

Y1 - 2010/6/1

N2 - Understanding the relationship between queueing delays and link utilization for general traffic conditions is an important open problem in networking research. Difficulties in understanding this relationship stem from the fact that it depends on the complex nature of arriving traffic and the problems associated with modelling such traffic. Existing AQM schemes achieve a "low delay" and "high utilization" by responding early to congestion without considering the exact relationship between delay and utilization. However, in the context of exploiting the delay/utilization tradeoff, the optimal choice of a queueing scheme's control parameter depends on the cost associated with the relative importance of queueing delay and utilization. The optimal choice of control parameter is the one that maximizes a benefit that can be defined as the difference between utilization and cost associated with queuing delay. We present two practical algorithms, Optimal Drop-Tail (ODT) and Optimal BLUE (OB), that are designed with a common performance goal: namely, maximizing this benefit. Their novelty lies in fact that they maximize the benefit in an online manner, without requiring knowledge of the traffic conditions, specific delay-utilization models, nor do they require complex parameter estimation. Packet level ns2 simulations are given to demonstrate the efficacy of the proposed algorithms and the framework in which they are designed.

AB - Understanding the relationship between queueing delays and link utilization for general traffic conditions is an important open problem in networking research. Difficulties in understanding this relationship stem from the fact that it depends on the complex nature of arriving traffic and the problems associated with modelling such traffic. Existing AQM schemes achieve a "low delay" and "high utilization" by responding early to congestion without considering the exact relationship between delay and utilization. However, in the context of exploiting the delay/utilization tradeoff, the optimal choice of a queueing scheme's control parameter depends on the cost associated with the relative importance of queueing delay and utilization. The optimal choice of control parameter is the one that maximizes a benefit that can be defined as the difference between utilization and cost associated with queuing delay. We present two practical algorithms, Optimal Drop-Tail (ODT) and Optimal BLUE (OB), that are designed with a common performance goal: namely, maximizing this benefit. Their novelty lies in fact that they maximize the benefit in an online manner, without requiring knowledge of the traffic conditions, specific delay-utilization models, nor do they require complex parameter estimation. Packet level ns2 simulations are given to demonstrate the efficacy of the proposed algorithms and the framework in which they are designed.

KW - Buffer management

KW - Delay/utilization tradeoff

KW - Internet routers

KW - Network utility

UR - http://www.scopus.com/inward/record.url?scp=77950864973&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=77950864973&partnerID=8YFLogxK

U2 - 10.1016/j.comcom.2010.02.014

DO - 10.1016/j.comcom.2010.02.014

M3 - Article

VL - 33

SP - 1108

EP - 1121

JO - Computer Communications

JF - Computer Communications

SN - 0140-3664

IS - 9

ER -