Thanks sir for your reply.

This is my understanding for quite a while and i hope i am not wrong for all
this time. So i'm happy that you agree.  This goes inline to the document
3750 QoS configuration examples - The document does not state this
objectively and I'm finding weird the fact of seeing lots of posts lately in
this forum of people adjusting priority queue size by tweaking the wrong
place so keen to hear any contrary intelligent opinions out there.

If anyone out there disagrees please join the discussion and don't be shy.



On Sat, Jul 24, 2010 at 12:49 PM, Randall Saborio <ill2...@gmail.com> wrote:

> Hi Daniel,
>
> I had to review again my notes and the documentation to tell for sure (wish
> I knew it out of my head as earlier I was studying a lot of the lan qos
> theory).
>
> You are correct, the srr-queue bandwidth shape means nothing when you
> configure the priority queue out. As it says on the doc:
> "All four queues participate in the SRR unless the expedite queue is
> enabled, in which case the first bandwidth weight is ignored and is not used
> in the ratio calculation. *The expedite queue is a priority queue, and it
> is serviced until empty* before the other queues are serviced. You enable
> the expedite queue by using the priority-queue out interface configuration
> command. "
>
> So what I get is the settings are ignore completely for the calculation of
> shared bandwidth for the other queues, and because the queue is serviced
> until empty.
>
> I'm all dizzy today from studying the LAN QoS and still can't say I know it
> all. :-/
>
>
> On Fri, Jul 23, 2010 at 5:49 PM, Daniel Berlinski <dberlin...@gmail.com>
> wrote:
> > Hello
> >
> > Can someone confirm my understanding.  The below question implies the use
> of
> > priority-queue out inteface command.
> >
> > For adjusting how much bandwidth is given to the egress priority queue of
> a
> > 3750/3560/2960 switch the interface command:
> > srr-queue bandwidth shape means nothing
> >
> > The srr command that tunes the buffer size of memory to be given to queue
> 1
> > is the one that will adjust the priority-queue depth required.
> >
> > Feedback please
> >
> >
> > _______________________________________________
> > For more information regarding industry leading CCIE Lab training, please
> > visit www.ipexpert.com
> >
> >
>
>
>
> --
> Randall "da ill" Saborio
> CCIE Voice Wannabe #10054675811
> (Real number coming this July 2010)
>
>
_______________________________________________
For more information regarding industry leading CCIE Lab training, please visit 
www.ipexpert.com

Reply via email to