From: Carlos G Mendioroz (tron@xxxxxxxxxxx)
Date: Mon May 13 2002 - 17:16:11 GMT-3
I've been looking at it and ... I don't know how it affects the shaping,
if it does.
The difference is that Bc ends up being equal to CIR when using traffic
rate.
Well, it is documented that the default Bc is CIR.
Also, that would lend Tc = 1s, but it is also documented that Tc will
never be outside the 10ms - 125ms range, so Tc ends up being 125ms.
What I don't really understand is if this does change the router
behaviour.
I don't think it will try to send 72000 bits in the first interval, and
in any case that does not matter that much. (also, being byte limit
2000, it can not be the case that it tries to send more than 16000 bits
in any interval).
As I understand it, it just should behave the same...
Nick Shah wrote:
>
> Guys, Hope if someone can shed some light on this...
>
> frame-relay traffic-rate can be used as a subset of specifying CIR/BE/BC
> parameters (in a map class). However different (bizarre !!) results are
> obtained by using it in lieu.
> for eg.
>
> map-class frame-relay MPCLASS
> frame-relay traffic-rate 64000 72000 (CIR, & PIR peak rate, peak rate has to
> be CIR+EIR)
>
> RouterA#sh traffic-shape
>
> Interface Se0
> Access Target Byte Sustain Excess Interval Increment Adapt
> VC List Rate Limit bits/int bits/int (ms) (bytes)
> Active
> 110 64000 2000 64000 8000 125 1000 -
> 120 64000 2000 64000 8000 125 1000 -
> 130 64000 2000 64000 8000 125 1000 -
>
> However ...
>
> map-class frame-relay MPCLASS
> frame-relay cir 64000
> frame-relay bc 8000
> frame-relay be 8000
>
> RouterA#sh traffic-shape
>
> Interface Se0
> Access Target Byte Sustain Excess Interval Increment Adapt
> VC List Rate Limit bits/int bits/int (ms) (bytes)
> Active
> 110 64000 2000 8000 8000 125 1000 -
> 120 64000 2000 8000 8000 125 1000 -
> 130 64000 2000 8000 8000 125 1000 -
> RouterA#
>
> However the important point to see is that the byte-limit still remains the
> same, but the Sustain bits/int (BC) changes. I know that BC = CIR/Tc (CIR
> per time interval), so does this mean that it will *burst* upto twice its
> capacity (64K ---- > 128K).
>
> Has anyone noticed this.. I remember John Neiberger pointing something
> similar out a few months ago, but I couldnt find any followup mail on that
> ...
>
> Nick
>
> Nick
This archive was generated by hypermail 2.1.4 : Thu Jun 13 2002 - 10:58:55 GMT-3