Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
158 tokens/sec
GPT-4o
7 tokens/sec
Gemini 2.5 Pro Pro
45 tokens/sec
o3 Pro
4 tokens/sec
GPT-4.1 Pro
38 tokens/sec
DeepSeek R1 via Azure Pro
28 tokens/sec
2000 character limit reached

ECN verbose mode: a statistical method for network path congestion estimation (0910.4455v2)

Published 23 Oct 2009 in cs.NI

Abstract: This article introduces a simple and effective methodology to determine the level of congestion in a network with an ECN-like marking scheme. The purpose of the ECN bit is to notify TCP sources of an imminent congestion in order to react before losses occur. However, ECN is a binary indicator which does not reflect the congestion level (i.e. the percentage of queued packets) of the bottleneck, thus preventing any adapted reaction. In this study, we use a counter in place of the traditional ECN marking scheme to assess the number of times a packet has crossed a congested router. Thanks to this simple counter, we drive a statistical analysis to accurately estimate the congestion level of each router on a network path. We detail in this paper an analytical method validated by some preliminary simulations which demonstrate the feasibility and the accuracy of the concept proposed. We conclude this paper with possible applications and expected future work.

Citations (7)

Summary

We haven't generated a summary for this paper yet.