Quality of Service 'shape average' command ?

[tedauction]

Hello, I have seem some QoS configs which include the 'shape average xxx' command. My question is what is the advantage of using this command if you WAN provider has given you a fixed speed link so there is no way of exceeding that anyway. Why not just do your bandwidth allocation percentages without using 'shape average xxx' ? Thanks kindly for any help.

Example of QoS policy using bandwidth percentages and 'shape average xxx'.

policy-map CHILD-WAN
class VOIP
priority percent 30
set ip dscp ef
class INTERACTIVE
bandwidth percent 35
fair-queue
class BULK
bandwidth percent 18
fair-queue
class NET-MGT
bandwidth percent 8
fair-queue
class class-default
bandwidth percent 9
fair-queue
policy-map PARENT-WAN
class ALL-TRAFFIC
shape average 9900000
service-policy CHILD-WAN

Then I have seen others that would also use bandwidth percentage allocation but without using the 'shape average xxx' command.

So why no 'shape average xxx' command here ?

e.g.

policy-map QUEUES
class Multi-Media
priority
class DataTrans
bandwidth remaining percent 70
class IA-High
bandwidth remaining percent 15
class class-default
bandwidth remaining percent 15

int xx

service-policy output QUEUES

[Philip D'Ath]

You have what is called a hierarchical QOS policy there.  Anytime the amount of bandwidth you have purchases is less than the access rate you need to do this.

If the bandwidth and access are the same (same 10Mb/s and 10Mb/s) then you do not need to do this.

[tedauction]

Hello thanks for that Philip, did you mean:

'If the bandwidth and access are the same (same 10Mb/s and 10Mb/s) then you "DO NOT NEED" to do this.

?

Thanks kindly.

[Philip D'Ath]

Oops, correct.  I have updated my original post.

[paul driver]

Hello

My understanding is Shaping an egress feature to allow maximum BW allocation of an physical ISP link which has a larger traffic rate than the committed interface rate agreed by you and the ISP.

The bandwidth command allows you to specify the minimum allocated rate allowed to traverse the link in times of congestion

So regards your QOS post, you have a shaped egress wan rate of 10MB and your class maps defined by the BW statements are allowed a minimum value of that 10MB in time of congestion

res

Paul

[Joseph W. Doherty]

Disclaimer

The Author of this posting offers the information contained within this posting without consideration and with the reader's understanding that there's no implied or expressed suitability or fitness for any purpose. Information provided is for informational purposes only and should not be construed as rendering professional advice of any kind. Usage of this posting's information is solely at reader's own risk.

Liability Disclaimer

In no event shall Author be liable for any damages wha2tsoever (including, without limitation, damages for loss of use, data or profit) arising out of the use or inability to use the posting's information even if Author has been advised of the possibility of such damage.

Posting

I believe the other posters have correctly answered, but for another "view", remember the bandwidth statements are used to apportion bandwidth when there's congestion.  If there's no congestion, they do not "engage".

An example of where this can be a problem, lets say you have a 100 Mbps hand-off, but have only contracted for 50 Mbps of bandwidth.  If you try to send 51 Mbps, your service provider will likely limit your bandwidth to 50 Mbps, but how do they manage that 50 Mbps?  Your QoS policy, which defines how you do want to manage such congestion, is not used because 51 Mbps does not cause any congestion on your 100 Mbps interface.

To insure the 51 Mbps is managed by your QoS, you use a shaper to artificially create congestion on your interface if there's more than 50 Mbps of traffic.  If there is, your QoS now manages that congestion as you desire.