[ippm] Re: FCFS Range (was RE: Re: Gorry Fairhurst's Discuss on draft-ietf-ippm-capacity-protocol-21: (with DISCUSS and COMMENT)=

[Gorry Fairhurst <[e-mail]>]

On 04/08/2025 16:26, [e-mail] wrote:
>
> Hi Gorry,
>
> to make some progress (may be stick with the sense, if you dislike the 
> way its written):
>
> For this reason, use of fixed rate flows is constrained to paths of a 
> domain under the control of the network operator performing the 
> measurement. This includes the case of a test by an operator of a 
> provider who is commercially responsible for a user perceiving his 
> contracted access bandwidth to be below expectation (this provider 
> might differ from the one operating the physical access in a 
> deregulated market).
>
> RG: I’d be happy to have less commercial terms, maybe my phantasy for 
> smart technical English statements are insufficient. Feeded by the 
> above and asked to remove commercial terms, an LLM came up with:
>
> Fixed rate testing is limited to paths within a domain managed by the 
> network operator carrying out the measurement. This also applies when 
> the operator tests a provider's system, where the provider is 
> accountable for a user's perception of having lower than expected 
> access bandwidth under their contract. Note that the testing provider 
> may differ from the physical access operator in a deregulated market.
>
> Suggestions / preferences are welcome.
>
This might be heading in a useful direction.

However: "Fixed rate testing is limited to paths within a domain managed 
by the network operator carrying out the measurement."

-- To me, "carrying out the measurement" is troublesome, because it 
doesn't the its the entity responsible for the service, and the point is 
if you since you don't have congestion control, please don't rate-test 
other operator network segments. If an operator subcontracts, that's 
still the same to me.


> Regards,
>
> Ruediger
>
Gorry
>
> *Von:*Gorry Fairhurst <[e-mail]>
> *Gesendet:* Montag, 4. August 2025 15:53
> *An:* Geib, Rüdiger <[e-mail]>; 
> [e-mail]; [e-mail]; 
> [e-mail]; [e-mail]
> *Cc:* [e-mail]
> *Betreff:* Re: AW: FCFS Range (was RE: [ippm] Re: Gorry Fairhurst's 
> Discuss on draft-ietf-ippm-capacity-protocol-21: (with DISCUSS and 
> COMMENT)=
>
> On 04/08/2025 11:35, [e-mail] wrote:
>
>     Hi Gorry,
>
>     Thanks. Sorry for the delay in replying.
>
>     “..scoping the fixed rate use to a domain under the control of the
>     network operator performing the measurement”:
>
>     In Germany at least, network providers may be obliged to offer:
>
>      1. L2 Access to other providers
>      2. L3 “Virtual Service” for resellers
>
>     And the regulator has an independent company testing all that,
>     including may rate testing. So I’d assume there to be some
>     interest in fixed rate testing on what is (re-)sold by the company
>     selling access to retail customers. Would this meet “domain under
>     control”?
>
>     I’d like to be clear about that before changing text. From an
>     operational point of view I understand an ops team desire swiftly
>     to measure what has been bought from a different provider on
>     wholesale terms for (re)assurance.
>
>     Regards,
>
>     Ruediger
>
> I was hoping this would include the networks which a "tester" has 
> (delegated) responsibility for the SLA, and exclude network segments 
> that simply were used (but not offered by them - such as within a CDN 
> or a peering to another operator's network).
>
> I would be happy to explore a different term to "domain under control" 
> if you wish to propose something, or to discuss an example if you 
> think we need one to be clear.
>
> Gorry
>
>     *Von:*Gorry Fairhurst <[e-mail]>
>     <mailto:[e-mail]>
>     *Gesendet:* Mittwoch, 9. Juli 2025 20:36
>     *An:* [e-mail]; [e-mail];
>     [e-mail]; Tommy Pauly
>     <[e-mail]> <mailto:[e-mail]>
>     *Cc:* [e-mail]
>     *Betreff:* Re: FCFS Range (was RE: [ippm] Re: Gorry Fairhurst's
>     Discuss on draft-ietf-ippm-capacity-protocol-21: (with DISCUSS and
>     COMMENT)=
>
>     I like the approach of getting the registry type and procedures clear.
>
>     I noted this question which needs a response:
>
>     On 04/07/2025 15:09, [e-mail] wrote:
>
>         The following text is in 4.1, can we update this and move this
>         to a new section after 3.1
>
>         [RG] Done. Please clarify, if the following sentence ending
>         2nd para of this new section is complete:
>
>         Measurements that generate traffic on shared paths (including
>         wifi and Internet paths)
>
>         need to consider the impact on other traffic. For this reason,
>         use of
>
>         fixed rate flows is constrained to paths
>
>         ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
>
>     I was hoping this could say what paths would be appropriate.
>
>     Perhaps suitable words could, for instance, scope the fixed rate
>     use to a domain under the control of the network operator
>     performing the measurement?
>
>     But, I am OK with another proposal that also helps clarify this.
>
>     Gorry
>