RE: Frame-Relay Traffic Shaping "Be" Value

From: Yasser Aly (blackyeyes00@hotmail.com)
Date: Fri Oct 24 2003 - 16:04:10 GMT-3


Dear All,

  Can someone elaborate and provide answer for the questions below ?
I searched for answers in the archives and couldn't find any.

Regards,
Yasser

>From: "Daniel Cisco Group Study" <danielcgs@imc.net.au>
>Reply-To: "Daniel Cisco Group Study" <danielcgs@imc.net.au>
>To: "Brian Dennis" <brian@labforge.com>, <ccielab@groupstudy.com>
>Subject: RE: Frame-Relay Traffic Shaping "Be" Value
>Date: Fri, 25 Apr 2003 15:14:05 +1000
>
>Thanks Brian.
>
>Makes a lot of sense.
>
>A couple of questions:
>
>(1) Say an 800 bit packet is ready to be transmitted and the bc bucket
>contains only 400 credits bits, and the be bucket contains say 2000 bits.
>Would it be correct to say that 400 bits would come out of the bc bucket,
>emptying that bucket, and 400 would come out of the be bucket, leaving only
>1600 bits? Your explanation implied that all the bits would come from the
>be bucket in this scenario. Which is correct? Maybe it wouldn't matter....
>
>(2) What happens when the bc value is set to a very small value, such that
>when a packet requests credits for transmission, it is requesting more than
>exists in the bc (and be) buckets? For example:
>
>map-class frame-relay SingleBucket
> frame-relay cir 16000
> frame-relay bc 2000
>
>Here, bc is only 250 bytes, and be=0. A 1500 byte packet cannot be
>serviced, as there will never be enough bc credits. Does some sort of debt
>scheme kick in? Do we then borrow credits from one or more future Tc's, to
>service the request?
>
>Daniel
>



This archive was generated by hypermail 2.1.4 : Mon Nov 24 2003 - 07:53:07 GMT-3