We study optimal path selection and bandwidth allocation algorithms for elastic flows under fairness constraints. We assume that traffic demands between origin-destination (O-D) pairs are given and are characterized by a minimum and a maximum bandwidth requirement. We consider the case when between each O-D pair there is a set of admissible paths through which the flows realizing the demand may be routed. (We say that a set of flows realizes a demand associated with an O-D pair, if the sum of the allocated bandwidths of these flows is at least this demand.) The allocation task is thus not only to determine the bandwidth of each flow realizing the demands, but also to identify the specific path for each demand such that a fairness specific utility is maximized. In the max-min fairness case this implies the maximization of the minimum allocated bandwidth, whereas in the proportional fairness case it implies the maximization of a logarithmic utility function. As an interesting (and practically relevant) generalization, we also allow multiple paths to realize a given demand (demand-split). We propose algorithms that solve the respective optimization tasks associated with the popular max-min and proportional fair sharing constraints. We demonstrate the efficiency and usefulness of these algorithms through numerical examples based on the backbone Polish public network. We believe that the results can be used in traffic engineering of networks carrying elastic traffic.
%0 Book Section
%1 Fodor2001667
%A Fodor, Gábor
%A Malicskó, Gábor
%A Pióro, Michal
%A Szymanski, Tomasz
%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 667 - 680
%R http://dx.doi.org/10.1016/S1388-3437(01)80160-8
%T Path optimization for elastic traffic under fairness constraints
%V 4
%X We study optimal path selection and bandwidth allocation algorithms for elastic flows under fairness constraints. We assume that traffic demands between origin-destination (O-D) pairs are given and are characterized by a minimum and a maximum bandwidth requirement. We consider the case when between each O-D pair there is a set of admissible paths through which the flows realizing the demand may be routed. (We say that a set of flows realizes a demand associated with an O-D pair, if the sum of the allocated bandwidths of these flows is at least this demand.) The allocation task is thus not only to determine the bandwidth of each flow realizing the demands, but also to identify the specific path for each demand such that a fairness specific utility is maximized. In the max-min fairness case this implies the maximization of the minimum allocated bandwidth, whereas in the proportional fairness case it implies the maximization of a logarithmic utility function. As an interesting (and practically relevant) generalization, we also allow multiple paths to realize a given demand (demand-split). We propose algorithms that solve the respective optimization tasks associated with the popular max-min and proportional fair sharing constraints. We demonstrate the efficiency and usefulness of these algorithms through numerical examples based on the backbone Polish public network. We believe that the results can be used in traffic engineering of networks carrying elastic traffic.
@incollection{Fodor2001667,
abstract = {We study optimal path selection and bandwidth allocation algorithms for elastic flows under fairness constraints. We assume that traffic demands between origin-destination (O-D) pairs are given and are characterized by a minimum and a maximum bandwidth requirement. We consider the case when between each O-D pair there is a set of admissible paths through which the flows realizing the demand may be routed. (We say that a set of flows realizes a demand associated with an O-D pair, if the sum of the allocated bandwidths of these flows is at least this demand.) The allocation task is thus not only to determine the bandwidth of each flow realizing the demands, but also to identify the specific path for each demand such that a fairness specific utility is maximized. In the max-min fairness case this implies the maximization of the minimum allocated bandwidth, whereas in the proportional fairness case it implies the maximization of a logarithmic utility function. As an interesting (and practically relevant) generalization, we also allow multiple paths to realize a given demand (demand-split). We propose algorithms that solve the respective optimization tasks associated with the popular max-min and proportional fair sharing constraints. We demonstrate the efficiency and usefulness of these algorithms through numerical examples based on the backbone Polish public network. We believe that the results can be used in traffic engineering of networks carrying elastic traffic. },
added-at = {2016-07-12T14:53:52.000+0200},
author = {Fodor, Gábor and Malicskó, Gábor and Pióro, Michal and Szymanski, Tomasz},
biburl = {https://www.bibsonomy.org/bibtex/2ee25bee945e89ad926f19110b26d19cd/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)80160-8},
editor = {Jorge Moreira de Souza, Nelson L.S. da Fonseca and de Souza e Silva, Edmundo A.},
interhash = {d7587f6c35a583be39783710e1ab6915},
intrahash = {ee25bee945e89ad926f19110b26d19cd},
issn = {1388-3437},
keywords = {itc itc17},
pages = {667 - 680},
publisher = {Elsevier},
series = {Teletraffic Science and Engineering },
timestamp = {2020-04-30T18:17:29.000+0200},
title = {Path optimization for elastic traffic under fairness constraints },
volume = 4,
year = 2001
}