Reasons not to deploy RED (1999) [84 citations — 2 self]
Popular Tags
No tags have been applied to this document.
BibTeX | Add To MetaCart
author = {Martin May and Jean Bolot and Christophe Diot and Bryan Lyles},
title = {Reasons not to deploy RED},
year = {1999}
}
Bookmarks
OpenURL
Abstract:
In this paper we question the benefits of RED by using a testbed made of two CISCO 7500 routers and up to 16 PCs to observe RED performance under a traffic load made of FTP transfers, together with HTTP traffic and non responsive UDP flows. The main results we found were, first, that RED with small buoeers does not improve significantly the performance of the network, in particular the overall throughput is smaller than with Tail Drop and the difference in delay is not significant. Second, parameter tuning in RED remains an inexact science, but has no big impact on the end-to-end performance. We argue that RED deployment is not straight forward, and we strongly recommend more research with realistic network settings to develop a full quantitative understanding of RED. Nevertheless, RED allows us to control the queue size with large buffers.
Citations
| 1738 | Random Early Detections Gateways for Congestion Avoidance – Floyd, Jacobson - 1993 |
| 7 | Dimitrios Stiliadis, Abhijit Choudhury ”Efficient Active Queue – Suter, Lakshman - 1997 |
| 2 | Random early detection gateways – Floyd - 1993 |
| 2 | Notes on using red for queue management and congestion avoidance. Nanog Workshop – Jacobson - 1998 |
| 1 | Recommendations on queue management and congestion avoidance – Crowcroft - 1997 |
| 1 | Ios conguration guide – Systems - 1998 |
