Re: [bmwg] Ben Campbell's No Objection on draft-ietf-bmwg-dcbench-terminology-17: (with COMMENT)

[Ben Campbell <[e-mail]>]

Thanks for the response. Comments inline:

Ben.

> On Jun 21, 2017, at 2:17 PM, Lucien <[e-mail]> wrote:
> 
> Hi Ben, 
> 
> Thanks for the comment! Please find my answers inline, and please acknowledge back!
> 
> Appreciate you taking the time to look at our work.
> 
> Lucien
> 
> On Wed, Jun 21, 2017 at 12:06 PM, Ben Campbell <[e-mail]> wrote:
> Ben Campbell has entered the following ballot position for
> draft-ietf-bmwg-dcbench-terminology-17: No Objection
> 
> When responding, please keep the subject line intact and reply to all
> email addresses included in the To and CC lines. (Feel free to cut this
> introductory paragraph, however.)
> 
> 
> Please refer to https://www.ietf.org/iesg/statement/discuss-criteria.html
> for more information about IESG DISCUSS and COMMENT positions.
> 
> 
> The document, along with other ballot positions, can be found here:
> https://datatracker.ietf.org/doc/draft-ietf-bmwg-dcbench-terminology/
> 
> 
> 
> ----------------------------------------------------------------------
> COMMENT:
> ----------------------------------------------------------------------
> 
> I find the naming of the draft fairly confusing. It goes way beyond
> "terminology"; it makes a number of normative (using 2119 language) statements
> about benchmarking procedures. I wonder why the sections about procedure did
> not go into the methodology draft instead. In general, I don't think putting
> normative language in an informational terminology draft is a good idea. (This
> would have been a DISCUSS, except that I am aware the bmwg has decided to make
> all its drafts informational and to still use 2119 language. For the record, I
> think that policy falls down with this draft.)
> 
> That's how we have decided it makes sense at BMWG to proceed with these two drafts. We have been hashing this out for 4+ years now and the current state is the consensus.  As author, I have no intention nor desire to change this

For the record, I think BMWG has adopted interpretations of both informational RFC and 2119 keywords that are at best unconventional, and at worst tortured. But I also recognize the precedent has been set, and don’t mean to hold this draft hostage to it. Thus it was a non-blocking comment which you can feel free to ignore :-)


> 
> I agree with the comment from others that this does not seem to be specific to
> datacenters.
> 
> Great, so did we, this is why we already worked on addressing this by:
> 	•  calling out specifically that this specifically applies to data center switches (defining what those are today)
> 	• stating clearly that it can be applied to switches out of the data center, but that's not the specific scope of this

Works for me, thanks.


> 
> - 2.2: Definitions of "store-and-forward" and "cut-through" when used in this
> context would be helpful. The first may be obvious, but the best I can do with
> "cut-through" is assume it means the opposite of "store-and-forward".
> 
> Those are cleary defined in RFC 1242, which we reference. We don't want to duplicate definitions here. 

Apologies, I missed the earlier citation to 1242, which was specifically in regard to these terms.

>   
>  
> 
> - 6.2: After reading the definition of "Incast" several times, I'm still not
> sure what it means or what is being measured.
> 
> It's many to one type of network traffic patterns. Its very commonly found in cloud data centers implementing distributed storage/compute frameworks (big data for example).   
> 

As written, it’s not clear to me if “Incast” means “one to many/many-to-many” communication in general, or the patterns of synchronization that result from that. The first paragraph seems to say the former but the second paragraph seems to say the latter.