Re: [aqm] adoption of draft-white-aqm-docsis-pie-01

[Mikael Abrahamsson <[e-mail]>]

On Sun, 29 Mar 2015, Szilveszter Nadas wrote:

> Hi,
>
> You have an interesting process to encourage mic comments. ;)
>
> Reviewed version: "draft-ietf-aqm-docsis-pie-00".

Me too.

Some suggestions and comments:

Abstract:

...
    "They are commonly positioned at the
    head of the bottleneck link for traffic in the upstream direction
    (from the customer),"

I find this sentence hard to read. I would suggest changes to the abstract 
along these lines (I will include _ around the changed parts)

"DOCSIS cable modems provide broadband Internet access to over one hundred 
million users worldwide.  They are commonly positioned _so that they 
handle the lowering speed adaptation in the upstream direction_ (from the 
customer) towards the Internet, and as a result, the impact of buffering 
and bufferbloat in the cable modem can have a significant effect on user 
experience.  The CableLabs DOCSIS specification _(introduced in)_ 3.1 
includes requirements for cable modems to support an Active Queue 
Management (AQM) algorithm that is intended to alleviate the impact that 
buffering has on latency sensitive traffic, while preserving bulk 
throughput performance.  In addition, the CableLabs DOCSIS 3.0 
specifications have also been amended to contain similar requirements."

I just feel that this makes it easier for a non-expert read by changing 
the second sentence to not talk about "head", and clarify that later that 
the AQM sections were introduced in 3.1 specification.

Section 1.

Here it says "CableLabs'" and in abstract it says "CableLabs" (without '). 
Is that correct?
I would recommend to add "upstream from the customer" in the second 
paragraph.

Section 3.

The list in 3 doesn't match the 3.1-3.4 headers, so my suggestion would be 
to get rid of the list in 3 and just keep the 3.1-3.4 headers.

Conclusion:

Then I read A, and that seems mostly to list changes to PIE for the 
DOCSIS-PIE implementation?

What is the aim of this document? To describe in an IETF environment what 
the DOCSIS-PIE implementation does? This is not stated in the abstract or 
introduction. I think it would be worthwhile to put in a few lines in 
either place to describe exactly what the aim of the document is, and 
potentially, what it isn't. Perhaps bump the overview from section 1 to 2 
and insert an introduction there?

It's my opinion that this is a document that is valuable to publish as an 
informational draft in IETF-AQM.


-- 
Mikael Abrahamsson    email: [e-mail]