The original paper is in English. Non-English content has been machine-translated and may contain typographical errors or mistranslations. ex. Some numerals are expressed as "XNUMX".
Copyrights notice
The original paper is in English. Non-English content has been machine-translated and may contain typographical errors or mistranslations. Copyrights notice
Le routage multivoies et la possibilité d'utiliser simultanément plusieurs chemins réseau sont proposés depuis longtemps comme moyens d'atteindre les objectifs de fiabilité et d'amélioration des performances d'un Internet de nouvelle génération. Cependant, son utilisation entraîne une livraison de paquets dans le désordre, ce qui est bien connu pour nuire aux performances TCP. Même si les protocoles de transport de nouvelle génération permettront sans aucun doute de mieux faire face à ce phénomène, un basculement complet vers ces nouveaux protocoles ne peut pas être effectué « du jour au lendemain » sur tous les appareils ; la réalité est que nous serons obligés de continuer à utiliser TCP sur de tels réseaux multivoies bien après que le déploiement du futur Internet soit terminé. Dans cet article, nous étudions l’utilisation de réorganisation des paquets au mieux -- un service de couche réseau facultatif pour améliorer les performances de toute session TCP en présence d'une livraison de paquets dans le désordre. Un tel service promet de permettre TCP non modifié tirer parti des gains de fiabilité et de performances offerts par un futur Internet multivoie sans subir les effets négatifs sur les performances généralement associés à la livraison de paquets dans le désordre. Nos expériences testent les performances de deux variantes TCP courantes sous dispersion de paquets avec différents nombres de chemins et quantités de variance de latence inter-chemins. Ils ont été réalisés à l'aide d'implémentations de réseau multivoies et de réorganisation de paquets mises en œuvre dans le banc d'essai d'Emulab. Nos résultats démontrent qu'un simple service de réorganisation au mieux peut isoler TCP du type de réorganisation que l'on pourrait attendre de l'utilisation de la dispersion des paquets sur des chemins disjoints dans un réseau étendu, et est capable de fournir des avantages significatifs en termes de performances avec peu d'inconvénients. -effets.
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Copier
John Russell LANE, Akihiro NAKAO, "On Best-Effort Packet Reordering for Mitigating the Effects of Out-of-Order Delivery on Unmodified TCP" in IEICE TRANSACTIONS on Communications,
vol. E93-B, no. 5, pp. 1095-1103, May 2010, doi: 10.1587/transcom.E93.B.1095.
Abstract: Multipath routing and the ability to simultaneously use multiple network paths has long been proposed as a means for meeting the reliability and performance improvement goals of a next generation Internet. However, its use causes out-of-order packet delivery, which is well known to hinder TCP performance. While next-generation transport protocols will no doubt better cope with this phenomenon, a complete switch to these new protocols cannot be made on all devices "overnight"; the reality is that we will be forced to continue using TCP on such multipath networks well after deployment of a future Internet is complete. In this paper, we investigate the use of best-effort packet reordering -- an optional network layer service for improving the performance of any TCP session in the presence of out-of-order packet delivery. Such a service holds the promise of allowing unmodified TCP to take advantage of the reliability and performance gains offered by a future multipath-enabled Internet without suffering the adverse performance effects commonly associated with out-of-order packet delivery. Our experiments test the performance of two common TCP variants under packet dispersion with differing numbers of paths and amounts of inter-path latency variance. They were conducted using multipath network and packet reorderer implementations implemented within the Emulab testbed. Our results demonstrate that a simple best-effort reordering service can insulate TCP from the type of reordering that might be expected from use of packet dispersion over disjoint paths in a wide-area network, and is capable of providing significant performance benefits with few ill side-effects.
URL: https://global.ieice.org/en_transactions/communications/10.1587/transcom.E93.B.1095/_p
Copier
@ARTICLE{e93-b_5_1095,
author={John Russell LANE, Akihiro NAKAO, },
journal={IEICE TRANSACTIONS on Communications},
title={On Best-Effort Packet Reordering for Mitigating the Effects of Out-of-Order Delivery on Unmodified TCP},
year={2010},
volume={E93-B},
number={5},
pages={1095-1103},
abstract={Multipath routing and the ability to simultaneously use multiple network paths has long been proposed as a means for meeting the reliability and performance improvement goals of a next generation Internet. However, its use causes out-of-order packet delivery, which is well known to hinder TCP performance. While next-generation transport protocols will no doubt better cope with this phenomenon, a complete switch to these new protocols cannot be made on all devices "overnight"; the reality is that we will be forced to continue using TCP on such multipath networks well after deployment of a future Internet is complete. In this paper, we investigate the use of best-effort packet reordering -- an optional network layer service for improving the performance of any TCP session in the presence of out-of-order packet delivery. Such a service holds the promise of allowing unmodified TCP to take advantage of the reliability and performance gains offered by a future multipath-enabled Internet without suffering the adverse performance effects commonly associated with out-of-order packet delivery. Our experiments test the performance of two common TCP variants under packet dispersion with differing numbers of paths and amounts of inter-path latency variance. They were conducted using multipath network and packet reorderer implementations implemented within the Emulab testbed. Our results demonstrate that a simple best-effort reordering service can insulate TCP from the type of reordering that might be expected from use of packet dispersion over disjoint paths in a wide-area network, and is capable of providing significant performance benefits with few ill side-effects.},
keywords={},
doi={10.1587/transcom.E93.B.1095},
ISSN={1745-1345},
month={May},}
Copier
TY - JOUR
TI - On Best-Effort Packet Reordering for Mitigating the Effects of Out-of-Order Delivery on Unmodified TCP
T2 - IEICE TRANSACTIONS on Communications
SP - 1095
EP - 1103
AU - John Russell LANE
AU - Akihiro NAKAO
PY - 2010
DO - 10.1587/transcom.E93.B.1095
JO - IEICE TRANSACTIONS on Communications
SN - 1745-1345
VL - E93-B
IS - 5
JA - IEICE TRANSACTIONS on Communications
Y1 - May 2010
AB - Multipath routing and the ability to simultaneously use multiple network paths has long been proposed as a means for meeting the reliability and performance improvement goals of a next generation Internet. However, its use causes out-of-order packet delivery, which is well known to hinder TCP performance. While next-generation transport protocols will no doubt better cope with this phenomenon, a complete switch to these new protocols cannot be made on all devices "overnight"; the reality is that we will be forced to continue using TCP on such multipath networks well after deployment of a future Internet is complete. In this paper, we investigate the use of best-effort packet reordering -- an optional network layer service for improving the performance of any TCP session in the presence of out-of-order packet delivery. Such a service holds the promise of allowing unmodified TCP to take advantage of the reliability and performance gains offered by a future multipath-enabled Internet without suffering the adverse performance effects commonly associated with out-of-order packet delivery. Our experiments test the performance of two common TCP variants under packet dispersion with differing numbers of paths and amounts of inter-path latency variance. They were conducted using multipath network and packet reorderer implementations implemented within the Emulab testbed. Our results demonstrate that a simple best-effort reordering service can insulate TCP from the type of reordering that might be expected from use of packet dispersion over disjoint paths in a wide-area network, and is capable of providing significant performance benefits with few ill side-effects.
ER -