[e2e] Queue size of routers

[Konstantina Papagiannaki]

> Long,
> 
> > > I believe (based solely on a single long-ago observation that an M/M/1/K
> > > queuing model seemed to predict the measured behaviour of core routers
> > > with short buffers pretty accurately) that the speculation in your third
> > > sentence above may in fact be true.
> > 
> > This sounds intriguing. Does anyone have a pointer to this work?
> 
> The observation I'm speaking of was buried in an internal report for a
> large network operator which studied what to spend money on to improve the
> network they had at the time.  This was a relatively long time ago, when
> "large" backbone networks were constructed from T3 circuits.  I left my
> copy of the report with the company almost 8 years ago.

There is some more recent work done in this area. Some pointer are the following:

C. Fraleigh and F. Tobagi and C. Diot (2003).
Provisioning IP Backbone Networks to Support Latency Sensitive Traffic.
In: IEEE Infocom. San Francisco. Mar 2003.

K. Papagiannaki and S. Moon and C. Fraleigh and P. Thiran and F. Tobagi and C. Diot (2002).
Analysis of Measured Single-Hop Delay from an Operational Backbone Network.
In: IEEE Infocom. New York. jun 2002.

The actual papers can be found at
http://ipmon.sprintlabs.com/pubs_trs/webpages/ATLPUB.php

I hope this helps.
Dina