I am studying Frame relay traffic shaping and have a question about how
the "frame-relay traffic-rate" command figures the Byte Limit as
displayed in the "show frame pvc xxx" or the "show traffic-shape"
commands. If I set "frame-relay traffic-rate 64000 64000", a "show
traffic-shape" shows a Byte Limit of 1000 which is essentially the size
of my token bucket. With no additional peak rate configured, the Be is
0 and Bc is 8000 bits/int, with tc = 125ms. So my byte limit is 8000/8
= 1000 bytes/int. Now if you change the command to "frame-relay
traffic-rate 64000 96000" the Byte Limit becomes 5000. This is where I
don't quite understand how it is coming up with that number. The cisco
docs say the following about this command:
"For example, entering the frame-relay traffic-rate 64000 96000 command
will result in a CIR of 64000 bps. Assuming 8 intervals of 125
milliseconds, the Bc is 64000/8 or 8000 bits. The Be value is
calculated by subtracting 64000 from 96000, so the one-second value is
32000 bits. For each 125-millisecond interval, the Be value is 4000
So if the Be is 4000 bits/int, that would be 500 bytes/int. If the
byte limit for a CIR of 64000 with no Be is 1000 bytes/int, then I
would think adding a Be of 4000 bits/int would make the Byte Limit 1500
rather than 5000.
All of the books I've read and Cisco's doc's seem to be in agreement
that the Be is 4000 bits/interval. So I don't understand how 4000
bits/int can translate to a 4000 byte/int increase in the Byte limit.
What am I missing?