Re: Vector of Profiles MIB

Hi Scott,

How are you?
See my embedded answers

Moti M

-----Original Message-----
From: [email protected] [mailto:[email protected]] On Behalf Of [email protected]
Sent: Monday, November 09, 2009 10:52 AM
To: [email protected]
Cc: [email protected]; Menachem Dodge
Subject: Re: [Adslmib] Vector of Profiles MIB

Hi Menachem and Markus,

I think we should have a discussion first about the
scope and purpose of the "Vector of Profiles MIB" proposal.

Can we please address some of the following issues :

1. Would the new MIB cover all xDSL technologies like the
    VDSL2 MIB does ?
[MM] Yes. It refers to the same TR-129 and G.997.1 documents

2. Would the new MIB cover all the parameters defined in G.997.1 ?
[MM] No. It more-or-less arranges the configuration parameters in a different way.
However, it does not address the status parameters, the PM parameters, alarm 
thresholds or notifications.

3. Would the only difference between the VDSL2 MIB and the new
    MIB be the way templates are represented ?
[MM] In addition to (2) above, the main issue is the way 'templates' are represented. 
The other issue is that the 'template', which is now called 'VoP', may belong to a 
single line. If that ('direct') mode is being utilized then the 'template' (VoP) is 
indexed by the line's ifIndex rather than by an ordered identifier/number.    

4. Can you state the things that the new MIB will have that the VDSL2
    MIB does not have.
[MM] In principle (ignoring few details) the configuration VoP is the same as what 
we call 'template'. The differences are: The 'direct' vs. 'indirect' VoP attachment 
methods, and better (with some compromises) division of the configuration parameters 
into smaller, focused, profiles. There are also few minor errors (in the vdsl2 MIB) that
are corrected in the new MIB.

5. Can you state the things that the VDSL2 MIB has that the new
    MIB does not have.
[MM] First of all, we're talking about the configuration part of the vdsl2 MIB. 
Other parts remain the same. I think the new MIB preserves the characteristics
of the vdsl2 MIB, except for the indexing. The VoP and related profiles indices are 
numeric rather than textual. 

6. Is the new MIB intended for embedded use such as in DSLAMs
    and xDSL line cards ?
[MM] yes.

7. Can you justify having two MIBs that do basically the same thing
    but do it in a slightly different way ?
[MM] There is no justification. The reason VoP was invented is because operators felt that
TR-129 model does not divide the configuration parameters in an optimal way and that may cause
waste of memory. The issue of 'direct' vs. 'indirect' attachment methods could also be solved 
with the TR-129 model. Was it pleasant for the BBF to publish two management models within 
a year or two? Certainly not.
   
8. If you are a developer of xDSL equipment, what criterion would you
    use to select one MIB or the other ?
[MM] A very good question. There's no natural mechanism to determine, such as different ifIndex values.


Regards,

Scott.


---- [email protected] wrote: 
> on behalf of Deutsche Telekom, I am interested on that item.
>  
> Thanks
> Markus
>  
> ________________________________
> 
> Von: [email protected] [mailto:[email protected]] Im
> Auftrag von Menachem Dodge
> Gesendet: Montag, 9. November 2009 09:22
> An: Menachem Dodge; '[email protected]'
> Betreff: Re: [Adslmib] Vector of Profiles MIB
> 
> 
> 
> Hello,
> 
>  
> 
> I haven't received any response to my email of last Monday. 
> 
>  
> 
> If anyone at all, has an interest in this item of work, please speak up
> now.
> 
>  
> 
> Best Regards,
> 
> Menachem Dodge
> 
>  
> 
> From: [email protected] [mailto:[email protected]] On
> Behalf Of Menachem Dodge
> Sent: Monday, November 02, 2009 7:07 PM
> To: '[email protected]'
> Subject: Re: [Adslmib] Vector of Profiles MIB
> 
>  
> 
> Hello All,
> 
>  
> 
> In April this year, several emails were exchanged regarding development
> of a "Vector of Profile" MIB in accordance with the Broadband Forum
> TR-165 Vector of Profiles document.
> 
> See the attached thread.
> 
>  
> 
> As this issue has been raised again, I would like to see if there is an
> interest amongst the members of the Working Group in this work.
> 
>  
> 
> Please share your thoughts on this issue.
> 
>  
> 
> Best Regards,
> 
> Menachem
> 

_______________________________________________
Adslmib mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/adslmib