Re: [RPRWG] Decoupling the service classes
When looking at the May 2003 meeting proceedings on the RPRWG web
site, it contains only 4 presentations.
Fom a look at the agenda slides and from the comments in David's mail,
I get a feeling there should have been more. Where are they?..
Best Regards,
Fredrik
---------------------------------------------
Fredrik Davik
Phone: +47 67 82 83 88
Mobile: +47 45 24 91 88
Fax: +47 67 82 82 01
Switchboard: +47 67 82 82 00
mailto:bjornfd@xxxxxxxxx
http://heim.ifi.uio.no/~bjornfd
http://www.simula.no/people_one.php?people_id=22
Simula Research Laboratory
P.O.Box 134, Lysaker
N-1325 Lysaker
Norway
"David V James" <dvj@xxxxxxxxxxxx> writes:
> All,
>
> Presentations at last week's meeting indicated a shared
> (by at least a few of us) concern towards the interactions
> between classB/C and classA traffic guarantees.
>
> To assist with thoughts in this area, I have attached a
> short description that (I believe) shows the "tough"
> scenario and describes how it could be handled. Its
> an overview only, but a useful starting point before
> jumping into the equations&simulations swamp.
>
> The key points are as follows:
> a) Downstream shaper scoping:
> 1) Must throttle classB&classC (not just classC).
> 2) Must throttle STQ, as well as add traffic.
> b) Downstream shaper rates:
> 1) If lightly congested downstream, the throttle
> limits would support downstream subclassA0 rates.
> 2) If highly congested downstream, the throttle
> limits would support downstream classA rates
> (subclassA0 plus subclassA1).
> c) ClassB should have precedence over classC.
>
> As is typical with prioritized classes, the difficult
> problems are not fairness within a class, but sustaining
> fairness while supporting prioritization between classes.
>
> DVJ
>
>
> David V. James
> 3180 South Ct
> Palo Alto, CA 94306
> Home: +1.650.494.0926
> +1.650.856.9801
> Cell: +1.650.954.6906
> Fax: +1.360.242.5508
> Base: dvj@xxxxxxxxxxxx