On Wednesday 05 January 2005 11:59, Jonathan Day wrote: <snip> > On the other hand, if it's an exact split over a > fairly long timeslice, you use a class-based queueing > system and measure what's been sent out of each queue. > You then predict what the net bandwidth is over the > whole timeslice, by looking at what gets sent and what > gets dropped. Each time you adjust the prediction, you > adjust the hard limits for the queues to allow out > whatever is left of that class' net bandwidth. I wish there was something like that available for PPP over ATM links. My ADSL connection varies between 160Kbps and 224Kbps depending on how many ATM cells are wasted by mostly null cells. I'm stuck with a rate of 160Kbps since that's the lowest stable rate I could find that works under the worst case scenario of many, small Ethernet packets. Sadly I miss out when my connection is not in that worst state. _______________________________________________ LARTC mailing list / LARTC@xxxxxxxxxxxxxxx http://mailman.ds9a.nl/mailman/listinfo/lartc HOWTO: http://lartc.org/
