Scott,
>
> If you use CAR to set the IP precedence at ingress, you can then use (say)
> dWFQ to utilize those precedence bits in weighting various flows. That
> should get you the traffic shaping you desire, as dWFQ controls how much
> bandwidth each flow can use during congestion.
As I understand it, dFWQ doesn't allow me *limit* bandwidth unless
there is contention for the egress route. IE it doesn't help me
say "this server only gets 1Mb/s maximum even if there is
no congestion in the network" - even if the precedence field is
set lower than any other traffic, it will still be transmitted
when the network is not congested. "All" dFWQ will give me is
the ability to preferentially send traffic within contract over
that out of contract.
[ Why do I want to restrict bandwidth when my network isn't
congested, and if I know noone else is using it? Because I'm
trying to put a meaningful traffic contract in place for this
type of server. If I don't limit it, those who buy 64kb/s
services at a 64kb/s price have no incentive to upgrade if they
get 10Mb/s performance in any case. But I don't want to break
TCP/IP performance by dropping on CAR excess traffic w/o
buffering ]
-- Alex Bligh GX Networks (formerly Xara Networks)
This archive was generated by hypermail 2b29 : Sun Aug 04 2002 - 04:13:13 EDT