From mboxrd@z Thu Jan 1 00:00:00 1970 From: Randolph Carter Date: Fri, 10 Oct 2003 19:59:38 +0000 Subject: Re: [LARTC] Bandwidth bounding in a Shared Channel (reused bandwidth) Message-Id: List-Id: References: In-Reply-To: MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: lartc@vger.kernel.org Sure stef, no aproximation methods? some probabilistic/statistical approach to the problem? "The two most common things in the universe are hydrogen and stupidity." On Fri, 10 Oct 2003, Stef Coene wrote: > On Friday 10 October 2003 16:54, Randolph Carter wrote: > > Sometimes (especially with DSL) in my location when you bought a channel > > say 128Kbps, there are some "peak" times when you can get even 256Kbps, > > and another times the channel can go down to least than 128Kbps(in that > > case the queuing discipline in the box would be doin't but nothing, and > > the queue would fall into the DSL router). When I used TC + HTB the users > > complain the speed has dramaticly decreased, and somehow they are right, > > the channel are not using the peaks(call bursts if you want). How can I do > > the queing discipline be aware of those changes, any idea; any > > suggestion? > This is allmot impossible to implement. You can not 'sense' the speed of the > link on your shaping box in real time. And that's what you need. > > Stef > > -- > stef.coene@docum.org > "Using Linux as bandwidth manager" > http://www.docum.org/ > #lartc @ irc.openprojects.net > _______________________________________________ LARTC mailing list / LARTC@mailman.ds9a.nl http://mailman.ds9a.nl/mailman/listinfo/lartc HOWTO: http://lartc.org/