[Tmrg] Now, where were we...?
d.wischik at cs.ucl.ac.uk (Damon Wischik) Thu, 19 November 2009 23:33 UTC
From: "d.wischik at cs.ucl.ac.uk"
Date: Thu, 19 Nov 2009 23:33:47 +0000
Subject: [Tmrg] Now, where were we...?
In-Reply-To: <BE0E1358-7C27-46A8-AF1E-D8D7CC834A52@ifi.uio.no>
References: <BE0E1358-7C27-46A8-AF1E-D8D7CC834A52@ifi.uio.no>
Message-ID: <4B05D5DB.3060708@cs.ucl.ac.uk>
Michael Welzl wrote: > Can delay ever be worse as a congestion indicator than > loss is? What kinds of misinterpretations can we have, > if we carefully interpret it? Delay is not a reliable indicator in the case of large multiplexers with small buffers. For example, Appenzeller et al. suggested that buffer size should be bandwidth*delay/sqrt(num flows). Under this recommendation, if the number of flows is very large then the maximum possible queueing delay is very small, and it may be small enough to be swamped by things like timer granularity. Packet loss should simply be (x-C)/x where x is the total load and C is the link speed, assuming that there are many flows and they are desynchronized, and this should continue to be a reliable indicator of congestion no matter how many flows there are. What I've written only applies to large multiplexers with small buffers. It doesn't apply to access links, where I can well believe that delay is an appropriate congestion indicator. Damon.
- [Tmrg] Now, where were we...? Lachlan Andrew
- [Tmrg] Now, where were we...? Larry Dunn
- [Tmrg] Now, where were we...? Michael Welzl
- [Tmrg] Now, where were we...? Stefan Hirschmann
- [Tmrg] Now, where were we...? Michael Welzl
- [Tmrg] Now, where were we...? Lachlan Andrew
- [Tmrg] Now, where were we...? Damon Wischik
- [Tmrg] Now, where were we...? Lachlan Andrew
- [Tmrg] Now, where were we...? John Heffner
- [Tmrg] Now, where were we...? Lachlan Andrew
- [Tmrg] Now, where were we...? Michael Welzl
- [Tmrg] Now, where were we...? Lachlan Andrew