Re: [aqm] WGLC on draft-ietf-aqm-eval-guidelines

[Roland Bless <[e-mail]>]

Hi Wes,

On 19.08.2015 at 04:52 Wesley Eddy wrote:
> There were only a couple of the "major issues" that I thought I should
> comment on as a co-chair of the WG:
> 
> 
>> 3) the overall number of tests and parameter combinations is really
>>   high
> 
> Are there particular permutations (or classes of permutations) that you
> can suggest to remove?  There's a balancing act between including enough
> to satisfy people that want to find edge cases and thoroughly
> characterize an algorithm, and the desire for a more easily tractable
> suite of tests.

Right now, I can't, it was merely an observation...

>> 4) from the discussed end-to-end metrics only latency/goodput metrics
>>   are used in the scenarios and for some of the scenarios these metrics
>>   are not suitable to show the desired behavior
> 
> It would be easier for the editors to improve this if you could suggest
> specific metrics to add to specific scenarios, I think.

We included some considerations in the comments to the individual
sections already. As a general notice, The document could benefit
by cross-referencing the scenarios against the documents referenced in
Major Issue 6.  Regarding particular metrics, suggesting one
would require exact understanding of particular test goals, proper
argumentation, and thus more time than LC deadline allowed.

Regards,
 Roland