[nsp] MSFC2 Rate Limiting

[Sukumar Subburayan]

In Cat6k you can user policing to rate-limit each MSFC vlan and it is done in
HW.

Say for eg, you have a customer on vlan 10 and want to restrict them to a 10
MB, here is an eg (using Native IOS):


a. enable 'qos globally' using 'mls qos'

b. You first create a class-map to attach them to a service class something
like:


class-map match-all COLO_CUSTOMER_RATELIMIT
  description Colo customer rate-limit
  match access-group name ALL-TRAFFIC

c. match all that customer's traffic using an ACL

ip access-list extended ALL-TRAFFIC
 permit ip any any

d. create a policy map which rate-limits that traffic to 10 MB

policy-map 10MB-RATELIMIT
 class COLO_CUSTOMER_RATELIMIT
     police 10240000 10240000 10240000 conform-action transmit exceed-action
			drop

e. apply the above policy as a input service-policy to the interface:

router#show run int vlan 10
Building configuration...

Current configuration : 270 bytes
!
interface Vlan10

 ip address 10.10.10.1 255.255.255.240
 service-policy input 10MB-RATELIMIT
end

Documentation is available at:

http://www.cisco.com/univercd/cc/td/doc/product/lan/cat6000/12_1e/swconfig/qos.htm#1019983

sukumar



I have tested the above and it works just fine. You can create different
policers for different customers..


On Jul 11,  5:06pm, Christopher J. Wolff wrote:
> Subject: [nsp] MSFC2 Rate Limiting
> Hello,
>
> I'm using a couple of 6509's w/MSFC2-PFC to aggregate colocation
> customers.  It 'seems' to be virtually impossible to rate-limit each
> MSFC vlan (via CAR), although I certainly hope this isn't the case.  If
> anyone has any workarounds thank you in advance.
>
> Christopher J. Wolff, VP CIO
> Broadband Laboratories, Inc.
> http://www.bblabs.com
>
> -----Original Message-----
> From: cisco-nsp-bounces at puck.nether.net
> [mailto:cisco-nsp-bounces at puck.nether.net] On Behalf Of Gert Doering
> Sent: Friday, July 11, 2003 1:10 PM
> To: Jason Lixfeld
> Cc: Temkin, David; cisco-nsp at puck.nether.net
> Subject: Re: [nsp] C7200 Bandwidth Points
>
> Hi,
>
> On Fri, Jul 11, 2003 at 03:12:53PM -0400, Jason Lixfeld wrote:
> > VXRs are godly!
>
> They still have a limit of 600 bandwidth points per bus, which can be
> reached quite quickly if you have no NPE-G1, two FastEs on the IO board,
> and some OC3 interfaces.
>
> I'm personally hoping for a "VXR++" chassis that has no "two busses"
> architecture anymore, but something like a crossbar fabric with a
> dedictated PCI "bus" per slot... saving PA investments (unlike the 7300)
> and still boosting the throughput.
>
> But I think this depends on whether Cisco sales or Cisco tech people
> decide.
>
> gert
> --
> USENET is *not* the non-clickable part of WWW!
>
> //www.muc.de/~gert/
> Gert Doering - Munich, Germany
> gert at greenie.muc.de
> fax: +49-89-35655025
> gert at net.informatik.tu-muenchen.de
> _______________________________________________
> cisco-nsp mailing list  cisco-nsp at puck.nether.net
> http://puck.nether.net/mailman/listinfo/cisco-nsp
> archive at http://puck.nether.net/pipermail/cisco-nsp/
>
> _______________________________________________
> cisco-nsp mailing list  cisco-nsp at puck.nether.net
> http://puck.nether.net/mailman/listinfo/cisco-nsp
> archive at http://puck.nether.net/pipermail/cisco-nsp/
>-- End of excerpt from Christopher J. Wolff