Preferential Treatment of Acknowledgment Packets in a Differentiated Services Network
In the context of networks offering Differentiated Services (DiffServ), we investigate the effect of acknowledgment treatment on the throughput of TCP connections. We carry out experiments on a testbed offering three classes of service (Premium, Assured and Best-Effort), and different levels of congestion on the data and acknowledgment path. We apply a full factorial statistical design and deduce that treatment of TCP data packets is not sufficient and that acknowledgment treatment on the reverse path is a necessary condition to reach the targeted performance in DiffServ efficiently. We find that the optimal marking strategy depends on the level of congestion on the reverse path. In the practical case where Internet Service Providers cannot obtain such information in order to mark acknowledgment packets, we show that the strategy leading to optimal overall performance is to copy the mark from the respective data packet into returned acknowledgement packets, provided that the affected service class is appropriately provisioned.
KeywordsData Packet Transmission Control Protocol Average Throughput Service Class Forward Path
Unable to display preview. Download preview PDF.
- 1.W. Almesberger Linux Network Traffic Control-Implementation Overview. http://icawww1.epfl.ch/linux-diffserv/, April 1999.
- 2.A. Basu and Z. Wang. A Comparative Study of Schemes for Differentiated Services. Technical report, Bell Laboratories, Lucent Technologies, July 1998.Google Scholar
- 3.S. Blake, D. Black, M. Carlson, E. Davies, Z.W ang, and W. Weiss. An Architecture for Differentiated Services. Request for Comments (Proposed Standard) 2475, Internet Engineering Task Force, October 1998.Google Scholar
- 4.K. Claffy. The Nature of the Beast: Recent Traffic Measurements from an Internet Backbone. In INET’98, 1998.Google Scholar
- 5.D. D. Clark and W. Fang. Explicit Allocation of Best-Effort Packet Delivery Service. ieanep, 6(4):362–373, August 1998.Google Scholar
- 7.S. Floyd and V. Jacobson. Link-sharing and resource management models for packet networks. IEEE/ACM Transactions on Networking, 3(4), August 1995.Google Scholar
- 8.Information Networks Division, Hewlett-Packard Company. Netperf: A Network Performance Benchmark. http://www.netperf.org, February 1995.
- 9.Raj Jain. The art of computer systems performance analysis: techniques for experimental design, measurement, simulation, and modeling. John Wiley, New York, 0-471-50336-3, 1991.Google Scholar
- 10.Peter W. M. John. Statistical Design and Analysis of Experiments. Society for Industrial and Applied Mathematics, 3600 University City Science Center, Philadelphia, PA 19104-2688, 0-89871-427-3, 1998.Google Scholar
- 11.S. Köhler and U. Schäfer. Performance Comparison of Different Class-and-Drop Treatment of Data and Acknowledgements in DiffServ IP Networks. Technical Report 237, University of Würzburg, August 1999.Google Scholar
- 12.T. V. Lakshman, U. Madhow, and B. Suter. TCP/IP Performance with Random Loss and Bidirectional Congestion. IEEE/ACM Transactions on Networking, 1998.Google Scholar
- 13.S. McCreary and K. Claffy. Trends in Wide Area IP Traffic Patterns-A View from Ames Internet Exchange. In ITC’00, Monterey, September 2000.Google Scholar
- 14.K. Nichols, V. Jacobson, and L. Zhang. A Two-bit Differentiated Services Architecture for the Internet. Internet Draft, Internet Engineering Task Force, May 1999. Work in progress.Google Scholar
- 15.K. Papagiannaki, P. Thiran, J. Crowcroft, and C. Diot. Preferential Treatment of Acknowledgment Packets in a Differentiated Services Network. Technical Report TR10-ATL-041001, Sprint Advanced Technology Laboratories, April 2001.Google Scholar