[Colloq] **Master Thesis Defense** Manav Khanna, Monday, June 3, 3pm

[Rachel Bates]

Date: :June 3rd, Monday, 3pm.

TITLE: Experimental Evaluation of Random Early Detection(RED) Gateways in
Heterogeneous Networks

Presenter: Manav Khanna
College of Computer Science, Northeastern University.

Advisor: V. Tsaoussidis
Reader: Ibrahim Matta

Abstract

We study the behavior of Random Early Detection (RED) in heterogeneous
environments and we report the impact of network heterogeneity on network
and application performance. Our testbed involves both wired and
wireless components, different transport protocols and Round Trip Times,
as well as a wide spectrum of channel conditions and flow contention which
cause packet corruption and congestion, respectively. Our evaluation
scenarios involve a non-monotonic transmission behavior of transport
protocols within one and the same congestion epoch.

Some of our results indicate that RED does not deal adequately with
heterogeneity.  We show that RED may (i) escalate further bandwidth
under-utilization by applying false congestion avoidance tactics;
(ii) degrade link utilization and fairness with diverse RTTs; and
(iii) penalize sophisticated protocols like TCP-SACK, which attempt to
recover more aggressively from wireless losses.  In summary, our
results call for further investigation of RED's efficiency in
heterogeneous networks and naturally raise the concern of RED's
compliance with the end-to-end argument.