Parallel TCP is one possible approach to increasing throughput of data transfer in Long Fat Networks (LFNs). Using parallel TCP is something of black art. As high-speed transport-layer protocols appear, e.g. HSTCP, it is necessary to reinvestigate the performance of parallel TCP, because a choice has to be make among them for the system. In this paper, the performance of parallel TCP is evaluated by mathematical analysis based on a simple dumbbell topology. Packet drop rate and aggregate goodput are used as two metrics to characterize the performance of parallel TCP. Two cases, namely synchronization and non-synchronization, are analyzed in detail when DropTail is deployed on routers. The synchronization case is common in using parallel TCP, but the goodput deteriorates seriously. The non-synchronization case may benefit parallel TCP, but extra mechanisms are required, and it is not easy to implement in the real world. The problem also remains even if Random Early Detection (RED) queue management is employed on routers. The analysis results show the difficulty in using parallel TCP in practice.
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.
Copy
Zongsheng ZHANG, Go HASEGAWA, Masayuki MURATA, "Analysis Evaluation of Parallel TCP: Is It Really Effective for Long Fat Networks?" in IEICE TRANSACTIONS on Communications,
vol. E90-B, no. 3, pp. 559-568, March 2007, doi: 10.1093/ietcom/e90-b.3.559.
Abstract: Parallel TCP is one possible approach to increasing throughput of data transfer in Long Fat Networks (LFNs). Using parallel TCP is something of black art. As high-speed transport-layer protocols appear, e.g. HSTCP, it is necessary to reinvestigate the performance of parallel TCP, because a choice has to be make among them for the system. In this paper, the performance of parallel TCP is evaluated by mathematical analysis based on a simple dumbbell topology. Packet drop rate and aggregate goodput are used as two metrics to characterize the performance of parallel TCP. Two cases, namely synchronization and non-synchronization, are analyzed in detail when DropTail is deployed on routers. The synchronization case is common in using parallel TCP, but the goodput deteriorates seriously. The non-synchronization case may benefit parallel TCP, but extra mechanisms are required, and it is not easy to implement in the real world. The problem also remains even if Random Early Detection (RED) queue management is employed on routers. The analysis results show the difficulty in using parallel TCP in practice.
URL: https://global.ieice.org/en_transactions/communications/10.1093/ietcom/e90-b.3.559/_p
Copy
@ARTICLE{e90-b_3_559,
author={Zongsheng ZHANG, Go HASEGAWA, Masayuki MURATA, },
journal={IEICE TRANSACTIONS on Communications},
title={Analysis Evaluation of Parallel TCP: Is It Really Effective for Long Fat Networks?},
year={2007},
volume={E90-B},
number={3},
pages={559-568},
abstract={Parallel TCP is one possible approach to increasing throughput of data transfer in Long Fat Networks (LFNs). Using parallel TCP is something of black art. As high-speed transport-layer protocols appear, e.g. HSTCP, it is necessary to reinvestigate the performance of parallel TCP, because a choice has to be make among them for the system. In this paper, the performance of parallel TCP is evaluated by mathematical analysis based on a simple dumbbell topology. Packet drop rate and aggregate goodput are used as two metrics to characterize the performance of parallel TCP. Two cases, namely synchronization and non-synchronization, are analyzed in detail when DropTail is deployed on routers. The synchronization case is common in using parallel TCP, but the goodput deteriorates seriously. The non-synchronization case may benefit parallel TCP, but extra mechanisms are required, and it is not easy to implement in the real world. The problem also remains even if Random Early Detection (RED) queue management is employed on routers. The analysis results show the difficulty in using parallel TCP in practice.},
keywords={},
doi={10.1093/ietcom/e90-b.3.559},
ISSN={1745-1345},
month={March},}
Copy
TY - JOUR
TI - Analysis Evaluation of Parallel TCP: Is It Really Effective for Long Fat Networks?
T2 - IEICE TRANSACTIONS on Communications
SP - 559
EP - 568
AU - Zongsheng ZHANG
AU - Go HASEGAWA
AU - Masayuki MURATA
PY - 2007
DO - 10.1093/ietcom/e90-b.3.559
JO - IEICE TRANSACTIONS on Communications
SN - 1745-1345
VL - E90-B
IS - 3
JA - IEICE TRANSACTIONS on Communications
Y1 - March 2007
AB - Parallel TCP is one possible approach to increasing throughput of data transfer in Long Fat Networks (LFNs). Using parallel TCP is something of black art. As high-speed transport-layer protocols appear, e.g. HSTCP, it is necessary to reinvestigate the performance of parallel TCP, because a choice has to be make among them for the system. In this paper, the performance of parallel TCP is evaluated by mathematical analysis based on a simple dumbbell topology. Packet drop rate and aggregate goodput are used as two metrics to characterize the performance of parallel TCP. Two cases, namely synchronization and non-synchronization, are analyzed in detail when DropTail is deployed on routers. The synchronization case is common in using parallel TCP, but the goodput deteriorates seriously. The non-synchronization case may benefit parallel TCP, but extra mechanisms are required, and it is not easy to implement in the real world. The problem also remains even if Random Early Detection (RED) queue management is employed on routers. The analysis results show the difficulty in using parallel TCP in practice.
ER -