Generalized Processor Sharing queues with heteregeneous traffic classes

Sem Borst, M.R.H. Mandjes, Miranda van Uitert

Research output: Contribution to journalArticleAcademicpeer-review

9 Citations (Scopus)

Abstract

We consider a queue fed by a mixture of light-tailed and heavy-tailed traffic. The two traffic flows are served in accordance with the generalized processor sharing (GPS) discipline. GPS-based scheduling algorithms, such as weighted fair queueing, have emerged as an important mechanism for achieving service differentiation in integrated networks. We derive the asymptotic workload behavior of the light-tailed traffic flow under the assumption that its GPS weight is larger than its traffic intensity. The GPS mechanism ensures that the workload is bounded above by that in an isolated system with the light-tailed flow served in isolation at a constant rate equal to its GPS weight. We show that the workload distribution is in fact asymptotically equivalent to that in the isolated system, multiplied with a certain pre-factor, which accounts for the interaction with the heavy-tailed flow. Specifically, the pre-factor represents the probability that the heavy-tailed flow is backlogged long enough for the light-tailed flow to reach overflow. The results provide crucial qualitative insight in the typical overflow scenario.
Original languageEnglish
Pages (from-to)806-845
Number of pages40
JournalAdvances in applied probability
Volume35
DOIs
Publication statusPublished - 2003

Fingerprint

Processor Sharing
Queue
Traffic
Workload
Telecommunication traffic
Overflow
Traffic Flow
Service Differentiation
Scheduling algorithms
Asymptotically equivalent
Queueing
Rate Constant
Scheduling Algorithm
Isolation
Class
Scenarios
Interaction

Keywords

  • METIS-207021

Cite this

Borst, Sem ; Mandjes, M.R.H. ; van Uitert, Miranda. / Generalized Processor Sharing queues with heteregeneous traffic classes. In: Advances in applied probability. 2003 ; Vol. 35. pp. 806-845.
@article{dfd2884039a84e59b95c341a09f9f098,
title = "Generalized Processor Sharing queues with heteregeneous traffic classes",
abstract = "We consider a queue fed by a mixture of light-tailed and heavy-tailed traffic. The two traffic flows are served in accordance with the generalized processor sharing (GPS) discipline. GPS-based scheduling algorithms, such as weighted fair queueing, have emerged as an important mechanism for achieving service differentiation in integrated networks. We derive the asymptotic workload behavior of the light-tailed traffic flow under the assumption that its GPS weight is larger than its traffic intensity. The GPS mechanism ensures that the workload is bounded above by that in an isolated system with the light-tailed flow served in isolation at a constant rate equal to its GPS weight. We show that the workload distribution is in fact asymptotically equivalent to that in the isolated system, multiplied with a certain pre-factor, which accounts for the interaction with the heavy-tailed flow. Specifically, the pre-factor represents the probability that the heavy-tailed flow is backlogged long enough for the light-tailed flow to reach overflow. The results provide crucial qualitative insight in the typical overflow scenario.",
keywords = "METIS-207021",
author = "Sem Borst and M.R.H. Mandjes and {van Uitert}, Miranda",
year = "2003",
doi = "10.1239/aap/1059486830",
language = "English",
volume = "35",
pages = "806--845",
journal = "Advances in applied probability",
issn = "0001-8678",
publisher = "University of Sheffield",

}

Generalized Processor Sharing queues with heteregeneous traffic classes. / Borst, Sem; Mandjes, M.R.H.; van Uitert, Miranda.

In: Advances in applied probability, Vol. 35, 2003, p. 806-845.

Research output: Contribution to journalArticleAcademicpeer-review

TY - JOUR

T1 - Generalized Processor Sharing queues with heteregeneous traffic classes

AU - Borst, Sem

AU - Mandjes, M.R.H.

AU - van Uitert, Miranda

PY - 2003

Y1 - 2003

N2 - We consider a queue fed by a mixture of light-tailed and heavy-tailed traffic. The two traffic flows are served in accordance with the generalized processor sharing (GPS) discipline. GPS-based scheduling algorithms, such as weighted fair queueing, have emerged as an important mechanism for achieving service differentiation in integrated networks. We derive the asymptotic workload behavior of the light-tailed traffic flow under the assumption that its GPS weight is larger than its traffic intensity. The GPS mechanism ensures that the workload is bounded above by that in an isolated system with the light-tailed flow served in isolation at a constant rate equal to its GPS weight. We show that the workload distribution is in fact asymptotically equivalent to that in the isolated system, multiplied with a certain pre-factor, which accounts for the interaction with the heavy-tailed flow. Specifically, the pre-factor represents the probability that the heavy-tailed flow is backlogged long enough for the light-tailed flow to reach overflow. The results provide crucial qualitative insight in the typical overflow scenario.

AB - We consider a queue fed by a mixture of light-tailed and heavy-tailed traffic. The two traffic flows are served in accordance with the generalized processor sharing (GPS) discipline. GPS-based scheduling algorithms, such as weighted fair queueing, have emerged as an important mechanism for achieving service differentiation in integrated networks. We derive the asymptotic workload behavior of the light-tailed traffic flow under the assumption that its GPS weight is larger than its traffic intensity. The GPS mechanism ensures that the workload is bounded above by that in an isolated system with the light-tailed flow served in isolation at a constant rate equal to its GPS weight. We show that the workload distribution is in fact asymptotically equivalent to that in the isolated system, multiplied with a certain pre-factor, which accounts for the interaction with the heavy-tailed flow. Specifically, the pre-factor represents the probability that the heavy-tailed flow is backlogged long enough for the light-tailed flow to reach overflow. The results provide crucial qualitative insight in the typical overflow scenario.

KW - METIS-207021

U2 - 10.1239/aap/1059486830

DO - 10.1239/aap/1059486830

M3 - Article

VL - 35

SP - 806

EP - 845

JO - Advances in applied probability

JF - Advances in applied probability

SN - 0001-8678

ER -