Re: Q. Terribly confused about interface bandwidth used for QoS

From: sabrina pittarel (sabri_esame@yahoo.com)
Date: Sat Sep 09 2006 - 19:56:41 ART

Thank you Brian,
 I understand now, but let me take an example of a possible lab scenario
 
 Assuming I have a P2P link with the DCE side clocked at 64000 and I'm asked do guarantee 10% of the bandwitdh on both sides to a given flow X and *not* to use the percentage keyword What would be the configuration of R1 and R2?
  
    1/1 1/1
  R1 -------------------------- R2
                           clockrate 64000
 
 
 Solution1)
 R1
 ---
 interface s1/1
   bandwith 64
   service-policy output s1/1-pm-out
 
 policy-map s1/1-pm-out
   class X
     bandwith 6
 
 R2
  ---
  interface s1/1
   clockrate 64000
    bandwith 64
    service-policy output s1/1-pm-out
  
  policy-map s1/1-pm-out
    class X
      bandwith 6

 
 Solution2)
 R1
  ---
  interface s1/1
    service-policy output s1/1-pm-out
  
  policy-map s1/1-pm-out
    class X
      bandwith 154 <<<<< 10% of 1544
  
  R2
   ---
   interface s1/1
    clockrate 64000
     service-policy output s1/1-pm-out
   
   policy-map s1/1-pm-out
     class X
       bandwith 154
 
 Sabrina
 
----- Original Message ----
From: Brian McGahan <bmcgahan@internetworkexpert.com>
To: sabrina pittarel <sabri_esame@yahoo.com>; ccielab@groupstudy.com
Sent: Saturday, September 9, 2006 3:15:00 PM
Subject: RE: Q. Terribly confused about interface bandwidth used for QoS calculations - please help

    In a practical application any time that the link speed does not
match the maximum transmission speed you should set bandwidth value
accordingly. For example suppose you have a FastEthernet interface
connected to your service provider but they are policing your traffic to
5Mbps. If you issue the "bandwidth percent 75" command under the MQC
you would be configuring a bandwidth guarantee beyond your reservable
capacity. The same case would hold true for serial interface, atm
interfaces, etc. In the lab exam I wouldn't worry about setting
bandwidth values unless there was a specific reason to do so though.

HTH,

Brian McGahan, CCIE #8593
bmcgahan@internetworkexpert.com

Internetwork Expert, Inc.
http://www.InternetworkExpert.com
Toll Free: 877-224-8987 x 705
Outside US: 775-826-4344 x 705
24/7 Support: http://forum.internetworkexpert.com
Live Chat: http://www.internetworkexpert.com/chat/

> -----Original Message-----
> From: nobody@groupstudy.com [mailto:nobody@groupstudy.com] On Behalf
Of
> sabrina pittarel
> Sent: Saturday, September 09, 2006 4:06 PM
> To: ccielab@groupstudy.com
> Subject: Q. Terribly confused about interface bandwidth used for QoS
> calculations - please help
>
> Hi all,
> can somebody help me clarify what is the interface bandwith used for
QoS
> calculations in various interface types?
> * Ethernet interface:
>
> R1 ------------- R2
> fe e
>
> R1 will auto negotiate a 10M speed with R2.
> 10Mbps will be used on R1 and R2 for QoS calculations. No need for
> explicit interface configuration.
>
> * Serial P2P interface Clocked at 64000
>
> R1 ------------- R2
> clockrate 64000
>
> How the configured clockrate will affect reference bandwidth for QoS
> calculation?
>
> Should we configure "bandwidth 64" on R1 and R2 serial interfaces?
> In my understanding the clockrate value determines how fast we can
really
> transmit on the media, but I don't see QoS calculations use that value
as
> reference number. Should we then configure the bandwidth command?
> Similar question for routing protocols.
> Should we alter default interface OSPF cost/eigrp interface bandwitdh
in
> respect to the media clocking?
>
> Sabrina
>
>

This archive was generated by hypermail 2.1.4 : Sun Oct 01 2006 - 16:55:40 ART