Re: 3750G

hi scott,

can you pls explain the diff between the burst in rate-limit vs burst in CB
policing.

thanks
naman

On 1 April 2010 18:06, Scott Morris <smorris_at_ine.com> wrote:

> You may want to do a little research on the commands and how they actually
> measure things. Because "burst" in one is not equivalent to "burst" in the
> other.
>
>
>
>
> *Scott Morris*, CCIE*x4* (R&S/ISP-Dial/Security/Service Provider) #4713,
>
> CCDE #2009::D, JNCIE-M #153, JNCIS-ER, CISSP, et al.
>
> JNCI-M, JNCI-ER
>
> evil_at_ine.com
>
>
> Internetwork Expert, Inc.
>
> http://www.InternetworkExpert.com
>
> Toll Free: 877-224-8987
>
> Outside US: 775-826-4344
>
>
> Knowledge is power.
>
> Power corrupts.
>
> Study hard and be Eeeeviiiil......
>
>
>
> Shahid Ansari wrote:
>
> rate-limit is legacy commnad but gives more flexible options for Bc than
> Police command which is not more than 1 Mb.
> Policing vs Shaping is best topic to discuss rather than Rate-limit vs
> Police : x
>
> Please rate :P
>
> Thanks
> Shahid
>
>
>
> On Thu, Apr 1, 2010 at 11:51 PM, naman sharma <naman.prep_at_gmail.com> <naman.prep_at_gmail.com> wrote:
>
>
>
> thanks shahid, just to get more clarity. rate-limit is also another form of
> policing so how come the burst size is diff is that.
>
> naman
>
>
> On 1 April 2010 13:46, Shahid Ansari <shahid1357_at_gmail.com> <shahid1357_at_gmail.com> wrote:
>
>
>
> Formula is right ..
>
> 3750GSWITCH(config-pmap-c)#police 250000000 ?
> <8000-1000000> Normal burst bytes
>
> max it can go 1Mb :) only
>
> but If you try rate-limit
>
> int g0/1/10
> no switchport
> ip address x.x.x.x x.x.x.x
> rate-limit input 250000000 46875000 46875000 conform-action transmit
> exceed-action drop
>
>
> in my opinion rate-limit will do what you want ?
>
> Thanks
> Shahid
>
>
>
> On Thu, Apr 1, 2010 at 10:08 PM, naman sharma <naman.prep_at_gmail.com> <naman.prep_at_gmail.com>wrote:
>
>
>
> Shahid,
>
> Yes i have a L3 interface in which i would be doing ingress policing. so
> in case i need to police at 250 mb on a 1 gig link. what would be the
> formula. is the same formula posted on cisco website
>
> normal burst = configured rate * (1 byte)/(8 bits) * 1.5 seconds
>
> extended burst = 2 * normal burst
>
> also can you share your configuration and what do you think will 3750G be able to police at 250 mb.
>
> thanks
> naman
>
>
>
> On 1 April 2010 03:01, Shahid Ansari <shahid1357_at_gmail.com> <shahid1357_at_gmail.com> wrote:
>
>
>
> Nauman ,
>
> As per the calculation bc should be 46875000, which is not supported
> :),Max you can put it 1 Mb .
> If it is L3 Interface , did you tried with rate-limit ?
> In 3750G , I have done some bandwidth control for Ingress as well as
> egress traffic by using Police(Ingress) and Srr-queue (Egress) commands
>
>
>
> Thanks
> Shahid Ansari
>
>
>
>
> On Tue, Mar 30, 2010 at 10:20 PM, naman sharma <naman.prep_at_gmail.com> <naman.prep_at_gmail.com>wrote:
>
>
>
> Hi experts,
>
> I have a 3750G switch and i am doing ingress traffic policing as it
> only
> supports that.
>
> Now i want to police my traffic at 250Mb on a 1Gig link. The problem is
> that
> the max burst capacity supported by 3750G is 1MB. now i am little
> confused.
>
> unlike shaping the buckets are not filled at regular time intervals in
> policing but are replenished in comparison to when did the last packet
> came
> and the Bc(burst) is only the depth of the bucket in which tokens are
> filled. This is when i am using single bucket logic.
>
> So would i be able to achieve my policing at 250MB even if i put my max
> burst to 1 MB. Need some help here on the logic.
>
> Thanks
> Naman
>
>
> Blogs and organic groups at http://www.ccie.net
>
> _______________________________________________________________________
> Subscription information may be found at:http://www.groupstudy.com/list/CCIELab.html
>
> Blogs and organic groups at http://www.ccie.net
>
> _______________________________________________________________________
> Subscription information may be found at: http://www.groupstudy.com/list/CCIELab.html

Blogs and organic groups at http://www.ccie.net
Received on Fri Apr 02 2010 - 01:27:50 ART