We consider a set of fluid On-Off flows that share a common server of capacity c and a finite buffer B. The server capacity is divided using Generalized Processor Sharing scheduling discipline. Each flow has a minimum service rate guarantee that exceeds its long term average demand ?i. The buffer sharing is unrestricted as long as there is available space; if the buffer is full, the necessary amount of fluid from the most demanding flows is discarded. When the On periods are heavy-tailed, we show that the loss rate of a particular flow i is asymptotically equal to the loss rate in a reduced system with capacity c-Sj?i?j and buffer B, where this flow is served in isolation. In particular, each flow perceives to have the whole buffer B to itself. This insight provides a new guideline for efficiently engineering differentiated quality of service in integrated multimedia networks.
%0 Book Section
%1 Jelenkovic2001719
%A Jelenkovic, Predrag
%A Momcilovic, Petar
%B Teletraffic Engineering in the Internet EraProceedings of the International Teletraffic Congress - ITC-I7
%D 2001
%E Jorge Moreira de Souza, Nelson L.S. da Fonseca
%E de Souza e Silva, Edmundo A.
%I Elsevier
%K itc itc17
%P 719 - 730
%R http://dx.doi.org/10.1016/S1388-3437(01)80164-5
%T Network multiplexer with generalized processor sharing and heavy-tailed on-off flows
%V 4
%X We consider a set of fluid On-Off flows that share a common server of capacity c and a finite buffer B. The server capacity is divided using Generalized Processor Sharing scheduling discipline. Each flow has a minimum service rate guarantee that exceeds its long term average demand ?i. The buffer sharing is unrestricted as long as there is available space; if the buffer is full, the necessary amount of fluid from the most demanding flows is discarded. When the On periods are heavy-tailed, we show that the loss rate of a particular flow i is asymptotically equal to the loss rate in a reduced system with capacity c-Sj?i?j and buffer B, where this flow is served in isolation. In particular, each flow perceives to have the whole buffer B to itself. This insight provides a new guideline for efficiently engineering differentiated quality of service in integrated multimedia networks.
@incollection{Jelenkovic2001719,
abstract = {We consider a set of fluid On-Off flows that share a common server of capacity c and a finite buffer B. The server capacity is divided using Generalized Processor Sharing scheduling discipline. Each flow has a minimum service rate guarantee that exceeds its long term average demand ?i. The buffer sharing is unrestricted as long as there is available space; if the buffer is full, the necessary amount of fluid from the most demanding flows is discarded. When the On periods are heavy-tailed, we show that the loss rate of a particular flow i is asymptotically equal to the loss rate in a reduced system with capacity c-Sj?i?j and buffer B, where this flow is served in isolation. In particular, each flow perceives to have the whole buffer B to itself. This insight provides a new guideline for efficiently engineering differentiated quality of service in integrated multimedia networks. },
added-at = {2016-07-12T14:53:52.000+0200},
author = {Jelenkovic, Predrag and Momcilovic, Petar},
biburl = {https://www.bibsonomy.org/bibtex/2cffac8f6809478b2a7cdec0acf8613d1/itc},
booktitle = {Teletraffic Engineering in the Internet EraProceedings of the International Teletraffic Congress - ITC-I7},
doi = {http://dx.doi.org/10.1016/S1388-3437(01)80164-5},
editor = {Jorge Moreira de Souza, Nelson L.S. da Fonseca and de Souza e Silva, Edmundo A.},
interhash = {eed006ef3155a22296aaf96e24a1e116},
intrahash = {cffac8f6809478b2a7cdec0acf8613d1},
issn = {1388-3437},
keywords = {itc itc17},
pages = {719 - 730},
publisher = {Elsevier},
series = {Teletraffic Science and Engineering },
timestamp = {2020-04-30T18:17:29.000+0200},
title = {Network multiplexer with generalized processor sharing and heavy-tailed on-off flows },
volume = 4,
year = 2001
}