Re: Traffic shaping vs rate limit

From: Petr Lapukhov (petr@internetworkexpert.com)
Date: Mon Jul 14 2008 - 14:31:54 ART

John,
Keep in mind the following major difference between traffic-shaping and
rate-limiting

1) The goal of traffic shaping is hard-limiting of sending rate, while the
purpose of rate-limiting is usually admission control based on burst sizes.

2) Traffic shaping hard-forces "fixed" sending rate, absorbing and smoothing
incoming packet bursts. Rate-limit never smoothes but propagates incoming
bursts, up to the limit allowed.

3) Rate-limiting drops (or marks) exceeding traffic, while traffic shaping
queues and delays packets. This dropping behavior has serious implications
on TCP-based applications, like you would never see 256Kbps for TCP flows
with your configuration.

4) Traffic-shaping is much more resource-consuming procedure, than
rate-limiting.

5) Traffic-shaping is based on leaky-bucket algorithm (time-quantum based),
while rate-limiting uses token-bucket (continuous flow). Therefore
traffic-shaping changes timing behavior of the flow (jitter, delay) whilte
rate-limiting does not.

So when you are asked to limit the *sending* rate, it sounds like a task for
shaper. However, based on (3) and (4) in real world it's usually reasonable
to use rate-limiting, especially for high-speed flows (the problem is to
find the optimal burst size).

Essentially, the question you asked lacks additional information to chose
the preferred method. You may want to look at the following link for
information on differences between policing and shaping. Our blog also has
special posts dedicated to that ;)

http://www.cisco.com/en/US/tech/tk543/tk545/technologies_tech_note09186a00800a3a25.shtml 

-- 
Petr Lapukhov, CCIE #16379 (R&S/Security/SP/Voice)
petr@internetworkexpert.com

Internetwork Expert, Inc.
http://www.InternetworkExpert.com
Toll Free: 877-224-8987
Outside US: 775-826-4344
Online Community: http://www.IEOC.com
CCIE Blog: http://blog.internetworkexpert.com

2008/7/14 John <jgarrison1@austin.rr.com>:

> Lets take a very ambigouis question.  Make sure traffic on int x does not
> exceed 256k.
>
>
> Intx
>
> traffic shape rate 256000 32000 0
>
> VS
> intx
>
> rate-limit 256000 32000 32000 conform transmit exceed drop
>
> I'm pretty sure I'm missing something to me these are the same
>
>
> _______________________________________________________________________
> Subscription information may be found at:
> http://www.groupstudy.com/list/CCIELab.html

This archive was generated by hypermail 2.1.4 : Mon Aug 04 2008 - 06:11:55 ART