A Study on the Active Queue Management Scheme with TCP Traffic


The KIPS Transactions:PartC, Vol. 8, No. 3, pp. 345-350, Jun. 2001
10.3745/KIPSTC.2001.8.3.345,   PDF Download:

Abstract

Supporting TCP data traffic over IP network requires special mechanisms to improve throughput and fairness. Several schemes have been proposed such as DT (Drop Tail), Random Drop and RED (Random Early Detection). RED algorithm was presented with the objective of avoiding congestion and maintaining the average queue length in a region of low delay and high throughput. In the current TCP/IP environment, TCP sources reacts to dropped packets by entering in the slow start phase, but network utilization will drop drastically. By detecting congestion and notifying only a randomly selected fraction of connection, RED causes to the global synchronization and fairness problem. In this paper, we study the active queue management algorithm to improve performance. and introduce an algorithm to maintain its fairness. It is also alleviates the global synchronization problem. Finally, we compare performance of proposed method with RED and Drop Tail in terms of goodput, network utilization and fairness.


Statistics
Show / Hide Statistics

Statistics (Cumulative Counts from September 1st, 2017)
Multiple requests among the same browser session are counted as one view.
If you mouse over a chart, the values of data points will be shown.


Cite this article
[IEEE Style]
J. Y. Yang, P. J. Lee, J. H. Kim, "A Study on the Active Queue Management Scheme with TCP Traffic," The KIPS Transactions:PartC, vol. 8, no. 3, pp. 345-350, 2001. DOI: 10.3745/KIPSTC.2001.8.3.345.

[ACM Style]
Jin Young Yang, Pal Jin Lee, and Jong Hwa Kim. 2001. A Study on the Active Queue Management Scheme with TCP Traffic. The KIPS Transactions:PartC, 8, 3, (2001), 345-350. DOI: 10.3745/KIPSTC.2001.8.3.345.