From: Vincent Mashburn (vmashburn@fedex.com)
Date: Thu Oct 26 2006 - 12:44:31 ART
Policing is a bit tricky when it comes to calculations. Whereas shaping
uses the CIR = Bc / Tc (or equivalent variant depending on shape average
or shape peak), policing does not use this formula. This is because the
token bucket is not replenished based on a Tc like shaping is. Policing
token buckets are replenished based on the packet inter-arrival time
(which can be difficult to determine). However, policing uses a maximum
inter-arrival time of 1 sec, so that would be a worst case scenario.
Also, policer token buckets are based on bytes and shaping buckets are
based on bits. The only formula that I have found is:
Bc = ((Inter-packet arrival time) * police rate ) / 8 ; where Bc is the
number of tokens (in bytes) placed into the bucket each "interval".
So, in your case, unless you have a good idea of what the average
inter-packet arrival time is, I would use the worst case scenario of
Bc = 3MB/8 = 375k
I am not really sure how this varies for the dual-rate policer.
Vince Mashburn
Voice / Data Engineer
901-263-5072
CCVP, CCNP, CCDA,Network +
Cisco IP Telephony Support Specialist
Cisco IP Telephony Operations Specialist
-----Original Message-----
From: nobody@groupstudy.com [mailto:nobody@groupstudy.com] On Behalf Of
Andi Bennett
Sent: Thursday, October 26, 2006 9:20 AM
To: ccielab@groupstudy.com
Subject: Rate limiting - calculating bc and be.
Hello all.
I've seen a few threads on the archives on this, but still no
definative answer that looks concrete to me.
If we have the two scenarios
below I'd like to know the CCIE lab way of calculating this.
1. We need to
police outbound to 3MB and have a normal burst of 200K and then a
maximum
burst of 300K.
2. We need to simply rate limit outbound to 3MB. Again, what
are the bc / be values and how are they calculated?
I've seen many threads on
using CIR/8 * 1.5 etc,but am still confused as to what the correct lab
answer
should be.
Thanks,
Andy.
Send instant messages to your online friends
http://uk.messenger.yahoo.com
This archive was generated by hypermail 2.1.4 : Wed Nov 01 2006 - 07:29:06 ART