Early Review of draft-ietf-detnet-flow-information-model-10
review-ietf-detnet-flow-information-model-10-rtgdir-early-rogge-2020-06-17-00

Hi,

I was asked by the IETF to review draft-ietf-detnet-flow-information-model-10.

First, I liked the Terminology section... for someone (as me) who is not
familiar with the document and the WG, this helps to make sure everyone gets
the basics right. Defining "Source" and "Destination" might feel strange at
first glance, but this is much better than letting someone guess what kind of
source is meant.

Overall this draft gives me a good picture what is information are exchanged on
DetNet.

Some nitpicks for specific chapters:

Chapter 2:
How is "Destination" defined for multicast traffic.

Chapter 5.5:
Are the packets 5.5 (and other chapters) are referring to meant to be IP
packets (with limited MTU?), or are they just "packets of information" to
define the maximum size of a data burst?

Chapter 5.8/6.7:
a)
DnIngressStatus is an "enumeration" (so there can be more than one), but it
cannot "PartialFailed" ? b) Is "PartialFailed" meant to be used for redundancy
or for multicast (or both)?

Chapter 5.9/6.3:
What time interval is used to specify the requirements of a DetNet Flow.

"At least 1 million octet every second" is not as difficult as "At least one
octet every microsecond", but both are the same octets/second value. Similar
differences can appear for latency and loss.

Henning Rogge