ACM Home Page
Please provide us with feedback. Feedback
Improving the start-up behavior of a congestion control scheme for TCP
Full text PdfPdf (831 KB)
Source Applications, Technologies, Architectures, and Protocols for Computer Communication archive
Conference proceedings on Applications, technologies, architectures, and protocols for computer communications table of contents
Palo Alto, California, United States
Pages: 270 - 280  
Year of Publication: 1996
ISBN:0-89791-790-1
Also published in ...
Author
Janey C. Hoe  Laboratory for Computer Science, Massachusetts Institute of Technology
Sponsor
SIGCOMM: ACM Special Interest Group on Data Communication
Publisher
ACM  New York, NY, USA
Bibliometrics
Downloads (6 Weeks): 9,   Downloads (12 Months): 97,   Citation Count: 56
Additional Information:

abstract   references   cited by   index terms   peer to peer  

Tools and Actions: Request Permissions Request Permissions    Review this Article  
DOI Bookmark: Use this link to bookmark this Article: http://doi.acm.org/10.1145/248156.248180
What is a DOI?

ABSTRACT

Based on experiments conducted in a network simulator and over real networks, this paper proposes changes to the congestion control scheme in current TCP implementations to improve its behavior during the start-up period of a TCP connection.The scheme, which includes Slow-start, Fast Retransmit, and Fast Recovery algorithms, uses acknowledgments from a receiver to dynamically calculate reasonable operating values for a sender's TCP parameters governing when and how much a sender can pump into the network. During the start-up period, because a TCP sender starts with default parameters, it often ends up sending too many packets and too fast, leading to multiple losses of packets from the same window. This paper shows that recovery from losses during this start-up period is often unnecessarily time-consuming.In particular, using the current Fast Retransmit algorithm, when multiple packets in the same window are lost, only one of the packet losses may be recovered by each Fast Retransmit; the rest are often recovered by Slow-start after a usually lengthy retransmission timeout. Thus, this paper proposes changes to the Fast Retransmit algorithm so that it can quickly recover from multiple packet losses without waiting unnecessarily for the timeout. These changes, tested in the simulator and on the real networks, show significant performance improvements, especially for short TCP transfers. The paper also proposes other changes to help minimize the number of packets lost during the start-up period.


REFERENCES

Note: OCR errors may be found in this Reference List extracted from the full text article. ACM has opted to expose the complete List rather than only correct and linked references.

1
2
3
 
4
S. Floyd. TCP and successive fast retransmits. ftp://ftp.ee.lbl.gov/papers/fastretrans.ps, October 1994.
 
5
Sally Floyd and Van Jacobson. On traffic phase effects in packet-switched gateways, ftp://ftp.ee.lbl/papers/phase.ps.Z.
 
6
A. Heybey. The network simulator. Technical report, MIT, September 1990.
 
7
J. C. Hoe. Start-up dynamics of TCP's congestion control and avoidance schemes, 1995.
8
 
9
V. Jacobson. Modified TCP congestion avoidance algorithm, end2end-interest mailing list (Apr.), 1990.
 
10
Van Jacobson. Berkeley TCP evolution from 4.3-tahoe to 4.3-reno. In Proceedings of the Eighteenth Internet Engineering Task Force, page 365, 1990.
11
 
12
M. Mathis, J. Mahdavi, S. Floyd, and A. Romanow. TCP selective acknowledgment options, ftp://ftp.ietf, cnri.reston.va.us/internetdrafts/draft-ietf-tcplw-sack-00.txt, April 1996. (Internet Draft, work in progress).
13
 
14
J. Postel. Transmission control protocol. Request for Comments 793, DDN Network Information Center, SRI International, September 1981.
15
 
16
 
17
W. R. Stevens. TCP/IP Illustrated, volume 1. Addison- Wesley Publishing Company, 1994.
 
18

CITED BY  56
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 


Peer to Peer - Readers of this Article have also read: