new version of TE framework draft: Comments

[Jim Boyle <jboyle@Level3.net>]

Please take a chance to review our current framework document.  Like
Like planning a wedding, the longer framework drafts take, the larger they
are likely to become.  So please take some time to review our current
framework document (draft-ietf-tewg-framework-03.txt), I suspect a WG last
call will be made shortly.

Here's my comments:
- it's big.
- timescale of picoseconds? (nit)

then what I'll call riders: restoration, mp-lambda-s, diffserv, webserv.

I don't think these should drift into the details.  Sections 6.5.2-3 don't
so bad, they are long (maybe I just have no attention span).

6.6 - what's this?  Web servers, and the need to tie in application load
splitting into the planning / provisioning cycles?  Maybe just strike
this?

6.7 (Diffserv TE) is way too far into the details, besides perhaps a bit
biased to one approach, to wit:

 "In order to provide enhanced quality of service in a Diffserv domain,
   it is not enough to implement proper buffer management and scheduling
   mechanisms. In addition to buffer management and scheduling"

maybe it is?  Is this a universal truth, is there a reference, is there no
dissenting view in the community?  Can another look be made at 6.7 in
light that this is in a framework document?

regards,

Jim




On Sun, 1 Apr 2001, Jim Boyle wrote:

> 
> Everyone, please use the new mailing list, thanks.
> 
> Jim
> 
> ---------- Forwarded message ----------
> Date: Fri, 30 Mar 2001 18:15:56 -0800
> From: xiaoxipe <xiaoxipe@cse.msu.edu>
> To: te-wg@UU.NET
> Cc: xxiao@photuris.com, awduche@movaz.com, angela.chiu@celion.com,
>      indra.widjaja@fnc.fujitsu.com, anwar@lucent.com,
>      xiaoxipe <xiaoxipe@cse.msu.edu>
> Subject: new version of TE framework draft
> 
> A new version of the TE framework draft "draft-ietf-tewg-framework-03.txt"
> is attached.  We've addressed the issues raised on the mailing list during
> the last call.  A text file describing all the changes made is also attached.
> 
> In summary, we didn't see any significant issues during the previous last
> call. Therefore, there is no change of technical substance in this new version.
> Some re-wording is done in section 6: the original "requirements" are now
> re-worded as "recommendations".
> 
> Comments are welcome.  We would like to request that the WG issue a
> last call on it shortly.
> 
> Xi-Peng
> 
> 
>