TCP implementations and false Time Out detection in OBS networks

Xiang Yu, Chunming Qiao, Yong Liu

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

This paper compares Reno, New-Reno and Selective Acknowledgements (SACK), the three most common TCP implementations today in (future) optical burst switched (OBS) networks. In general, SACK, which considers multiple Triple Duplicated ACKed (TD) losses in one round, is found to perform best in OBS networks, while New-Reno, which improves Reno in packet switched networks by fast retransmission in responding to partial ACKs, may however perform worse than Reno. All three TCP implementations react to a Time Out (TO) loss in the same way (i.e. using Slow Start). In OBS networks, where a burst may contain all packets from one round, and a burst loss occurs mainly due to contention instead of buffer overflow, such a TO event may no longer imply heavy congestion, or in other words, it may be a false TO or FTO. Such FTOs, which may be common in OBS networks especially for fast TCP flows, can significantly degrade the performance of all existing TCP implementations. Accordingly, we also propose a new TCP implementation called Burst TCP (BTCP) which can detect FTOs and react properly, and as a result, improve over the existing TCP implementations significantly.

Original languageEnglish (US)
Title of host publicationProceedings - IEEE INFOCOM
Pages774-784
Number of pages11
Volume2
StatePublished - 2004
EventIEEE INFOCOM 2004 - Conference on Computer Communications - Twenty-Third Annual Joint Conference of the IEEE Computer and Communications Societies - Hongkong, China
Duration: Mar 7 2004Mar 11 2004

Other

OtherIEEE INFOCOM 2004 - Conference on Computer Communications - Twenty-Third Annual Joint Conference of the IEEE Computer and Communications Societies
CountryChina
CityHongkong
Period3/7/043/11/04

Fingerprint

Switching networks
Packet networks

ASJC Scopus subject areas

  • Electrical and Electronic Engineering
  • Hardware and Architecture

Cite this

Yu, X., Qiao, C., & Liu, Y. (2004). TCP implementations and false Time Out detection in OBS networks. In Proceedings - IEEE INFOCOM (Vol. 2, pp. 774-784)

TCP implementations and false Time Out detection in OBS networks. / Yu, Xiang; Qiao, Chunming; Liu, Yong.

Proceedings - IEEE INFOCOM. Vol. 2 2004. p. 774-784.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Yu, X, Qiao, C & Liu, Y 2004, TCP implementations and false Time Out detection in OBS networks. in Proceedings - IEEE INFOCOM. vol. 2, pp. 774-784, IEEE INFOCOM 2004 - Conference on Computer Communications - Twenty-Third Annual Joint Conference of the IEEE Computer and Communications Societies, Hongkong, China, 3/7/04.
Yu X, Qiao C, Liu Y. TCP implementations and false Time Out detection in OBS networks. In Proceedings - IEEE INFOCOM. Vol. 2. 2004. p. 774-784
Yu, Xiang ; Qiao, Chunming ; Liu, Yong. / TCP implementations and false Time Out detection in OBS networks. Proceedings - IEEE INFOCOM. Vol. 2 2004. pp. 774-784
@inproceedings{a0e7fda3cbeb497ea4e2d1f64ea01462,
title = "TCP implementations and false Time Out detection in OBS networks",
abstract = "This paper compares Reno, New-Reno and Selective Acknowledgements (SACK), the three most common TCP implementations today in (future) optical burst switched (OBS) networks. In general, SACK, which considers multiple Triple Duplicated ACKed (TD) losses in one round, is found to perform best in OBS networks, while New-Reno, which improves Reno in packet switched networks by fast retransmission in responding to partial ACKs, may however perform worse than Reno. All three TCP implementations react to a Time Out (TO) loss in the same way (i.e. using Slow Start). In OBS networks, where a burst may contain all packets from one round, and a burst loss occurs mainly due to contention instead of buffer overflow, such a TO event may no longer imply heavy congestion, or in other words, it may be a false TO or FTO. Such FTOs, which may be common in OBS networks especially for fast TCP flows, can significantly degrade the performance of all existing TCP implementations. Accordingly, we also propose a new TCP implementation called Burst TCP (BTCP) which can detect FTOs and react properly, and as a result, improve over the existing TCP implementations significantly.",
author = "Xiang Yu and Chunming Qiao and Yong Liu",
year = "2004",
language = "English (US)",
volume = "2",
pages = "774--784",
booktitle = "Proceedings - IEEE INFOCOM",

}

TY - GEN

T1 - TCP implementations and false Time Out detection in OBS networks

AU - Yu, Xiang

AU - Qiao, Chunming

AU - Liu, Yong

PY - 2004

Y1 - 2004

N2 - This paper compares Reno, New-Reno and Selective Acknowledgements (SACK), the three most common TCP implementations today in (future) optical burst switched (OBS) networks. In general, SACK, which considers multiple Triple Duplicated ACKed (TD) losses in one round, is found to perform best in OBS networks, while New-Reno, which improves Reno in packet switched networks by fast retransmission in responding to partial ACKs, may however perform worse than Reno. All three TCP implementations react to a Time Out (TO) loss in the same way (i.e. using Slow Start). In OBS networks, where a burst may contain all packets from one round, and a burst loss occurs mainly due to contention instead of buffer overflow, such a TO event may no longer imply heavy congestion, or in other words, it may be a false TO or FTO. Such FTOs, which may be common in OBS networks especially for fast TCP flows, can significantly degrade the performance of all existing TCP implementations. Accordingly, we also propose a new TCP implementation called Burst TCP (BTCP) which can detect FTOs and react properly, and as a result, improve over the existing TCP implementations significantly.

AB - This paper compares Reno, New-Reno and Selective Acknowledgements (SACK), the three most common TCP implementations today in (future) optical burst switched (OBS) networks. In general, SACK, which considers multiple Triple Duplicated ACKed (TD) losses in one round, is found to perform best in OBS networks, while New-Reno, which improves Reno in packet switched networks by fast retransmission in responding to partial ACKs, may however perform worse than Reno. All three TCP implementations react to a Time Out (TO) loss in the same way (i.e. using Slow Start). In OBS networks, where a burst may contain all packets from one round, and a burst loss occurs mainly due to contention instead of buffer overflow, such a TO event may no longer imply heavy congestion, or in other words, it may be a false TO or FTO. Such FTOs, which may be common in OBS networks especially for fast TCP flows, can significantly degrade the performance of all existing TCP implementations. Accordingly, we also propose a new TCP implementation called Burst TCP (BTCP) which can detect FTOs and react properly, and as a result, improve over the existing TCP implementations significantly.

UR - http://www.scopus.com/inward/record.url?scp=8344240421&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=8344240421&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:8344240421

VL - 2

SP - 774

EP - 784

BT - Proceedings - IEEE INFOCOM

ER -