Management of parallel UBR flows over TCP in congested ATM networks

Authors:
A. Ishtiaq;Y. Okabe;M. Kanazawa
Affiliations:
Graduate School of Informatics, Kyoto University, Sakyo-ku, Yoshidahonmachi, Kyoto 606-8501, Japan;Graduate School of Informatics, Kyoto University, Sakyo-ku, Yoshidahonmachi, Kyoto 606-8501, Japan;Graduate School of Informatics, Kyoto University, Sakyo-ku, Yoshidahonmachi, Kyoto 606-8501, Japan
Venue:
Computer Communications
Year:
2004

Citing 7
Cited 0

Simulation-based comparisons of Tahoe, Reno and SACK TCP

ACM SIGCOMM Computer Communication Review
Improving the start-up behavior of a congestion control scheme for TCP

Conference proceedings on Applications, technologies, architectures, and protocols for computer communications
Forward acknowledgement: refining TCP congestion control

Conference proceedings on Applications, technologies, architectures, and protocols for computer communications
TCP byte counting refinements

ACM SIGCOMM Computer Communication Review
TCP Boston: A Fragmentation-tolerant TCP Protocol for ATM Networks

TCP Boston: A Fragmentation-tolerant TCP Protocol for ATM Networks
A report on recent developments in TCP congestion control

IEEE Communications Magazine
Dynamics of TCP traffic over ATM networks

IEEE Journal on Selected Areas in Communications

Quantified Score

Hi-index	0.24

Visualization

Abstract

The Linux transmission control protocol (TCP) implementation consists of TCP Reno, NewReno and selective acknowledgement options. This has improved the throughput of TCP/IP over the Internet. However, TCP over plain ATM networks significantly suffers from throughput degradation. In this paper, we describe the throughput performance of Linux TC Pover parallel UBR flows in the presence of high priority traffic class like CBR. We then proposed a modified congestion control algorithm for TCP to improve its throughput over ATM networks. The experimental results are provided to claim that our proposed algorithm dynamically senses the congestion in the network and adjusts the sender rate more accurately. The throughput results are thoroughly better than other algorithms and at the same time the fairness and synchronization is achieved fairly better among contending TCP flows.