by Lorenzo Vicisano, Jon Crowcroft
Proceedings of the Third International Workshop on High Performance Protocol Architectures, HIPPARCH '97
http://www.cs.ucl.ac.uk/external/L.Vicisano/pubb/hipparch97.ps.gz
Add To MetaCart
Abstract:
In this paper we depict and evaluate the performance of a protocol for reliable bulkdata transfer from one sender to many receivers simultaneously, using the Internet multicast infrastructure (MBone). The protocol, featuring a TCP-friendly congestion control algorithm, has been designed aiming at achieving a complete scalability of the system with the respect to the number of receivers. For this reason both reliability and congestion control are carried out by receivers, avoiding to involve the sender in the congestion control feedback loop and relieving it from the burden of carrying out retransmission for all receivers. The receiver-driven congestion control algorithm presented is based on a redundant layered organisation of data, where redundancy is used to spread the information being sent into a large number of data units, providing receivers with flexibility in accepting some of them rather than others, and still being able to complete the reception. Reliability is based on a probabilistic approach and is achieved using redundancy and continuous transmission, providing graceful degradation of the protocol efficiency when the loss rate increases. Other contributions in this paper include: simulation of the protocol in competition with TCP with drop-tail and RED routers; analysis of the stable operating point; implementation and measurements of the protocol (future). 1
Citations
|
1790
|
Congestion avoidance and control
– Jacobson
- 1988
|
|
1362
|
RTP: A Transport Protocol for Real-Time Applications
– Schulzrinne, Casner, et al.
- 1996
|
|
573
|
Receiver-driven layered multicast
– McCanne, Jacobson, et al.
- 1996
|
|
342
|
TCP slow start, congestion avoidance, fast retransmit, and fast recovery algorithms
– Stevens
- 1997
|
|
292
|
Effective erasure codes for reliable computer communication protocols
– Rizzo
- 1997
|
|
189
|
Internet group management protocol, version 2,” Request for Comments (Proposed Standard) 2236, Internet Engineering Task Force
– Fenner
- 1997
|
|
119
|
Modified TCP Congestion Avoidance Algorithm,” end2end-interest mailing list (ftp://ftp.ee.lbl.gov/email/vanj.90apr30.txt
– Jacobson
- 1990
|
|
116
|
Router Mechanisms to Support End-to-End Congestion Control. ftp://ftp.ee.lbl.gov/papers/collapse.ps
– Fall, Floyd
- 1997
|
|
85
|
A reliable multicast data distribution protocol based on software FEC techniques
– Rizzo, Vicisano
- 1997
|
|
72
|
Network News Transfer Protocol: A Proposed Standard for the Stream-Based Transmission of News",RFC 977
– Kantor, Lapsley
- 1986
|
|
58
|
Internet cache protocol (ICP), version 2
– Wessels, Claffy
- 1997
|
|
49
|
Start-up Dynamics of TCP’s Congestion Control and Avoidance Schemes
– Hoe
- 1995
|
|
48
|
Extending TCL for dynamic object-oriented programming
– Wetherall, Lindblad
- 1995
|
|
21
|
The Stationary Distribution of Ideal TCP Congestion Avoidance
– OTT, KEMPERMAN, et al.
- 1996
|
|
16
|
Notes on a cumulative layered organisation of data packets across multiple streams with different rates. Research Note Note RN/98/25
– Vicisano
- 1998
|
|
6
|
R.Jain "Analysis of the Increase and Decrease Algorithm hms for Congestion Avoidance
– Chiu
- 1989
|
|
2
|
B-MART: Bulk-data (non-real-time) Multiparty Adaptive Reliable Transfer Protocol
– Vicisano, Handley, et al.
- 1997
|
|
1
|
Simulation Based Comparisons of Thaoe
– Fall
- 1996
|