[네트워크]Schduling to improve the performance of TCP and UDP flows

소개글

The Internet today carries different types of traffic that have different service requirements. A large fraction of the traffic is either Web traffic requiring low response time or peer-to-peer traffic requiring high throughput. Meeting both performance requirements in a network where routers use droptail or RED for buffer management and FIFO as the service policy is an elusive goal. It is therefore worthwhile to investigate alternative scheduling and buffer management policies to improve the response time of Web traffic.
In this paper, we propose RED queue whose simulation manifests better efficiency over tail-drop queue.

목차

1. introduction

2. FIFO scheduling
2,1 explanation of the NS script
2.2 2.2 FIFO Simulation Results

3. Trace analysis

4. RED queue
4.1 RED Scenario
4.2 RED Queue Trace Graph

5. Conclusion

6, Reference

본문내용

Active queue management has been proposed to support end-to-end congestion control in the Internet. The aim of active queue management is to anticipate and signal congestion before it actually occurs. One of the active queue management approaches is random early discard (RED). RED allows for dropping packets before the buffer overflows and was proposed to achieve various goals such as ensuring fairness among sources with different roundtrip times, reducing average transfer delays, punishing unresponsive flows, and ensuring better interaction with TCP congestion control. A flow is declared unresponsive if its source does not respond to network congestion. However, in
practice RED fails to offer most of these features. For example, RED cannot protect TCP flows against unresponsive flows. Similarly, for Web flows, first-in first-out (FIFO) with RED was shown to yield similar flow transfer times as FIFO with droptail.
FIFO scheduling is one of the most important mechanisms to provide service guarantees is scheduling.

참고 자료

[1] size-based scheduling to improve the performance of short TCP flows – Idris A. Rai, Ernst W. Niersack, and Guillaume Urvoy-Keller, Institut Eurecom
[2] Optical networks – Ramswami Sivarajan
[3] http://computer.org/intelligent
[4] http://www.linux.co.kr/
[5] http://www.nist.gov/dads/HTML/queue.html
[6] S. Floyd and V. Jacobson, “Random Early Detection Gateways for Congestion Avoidance,” IEEE/ACM Trans. Net., vol. 1, no.4 Aug 1993, pp. 397-413
[7] www.icir.org/floyd/red.html
[8] http://kelp.or.kr/korweblog/