Re: [ippm] Fwd: Call for adoption: IPPM Registry Drafts

[lingli deng <[e-mail]>]

hi vero,

i would be happy to be involved.
as expressed earlier, i believe there are cases where passive measurement
is expected to be more helpful.

lingli

Vero Zheng <[e-mail]>于2014年3月13日星期四写道：
>> Having no passive measurements framework document yet, my feeling is that
>> the proposed time schedule is pretty tight. The content of a
>> (future) passive measurement RFC will likely influence on the registry
structure.
>> I might be desirable to start work on the passive measurement framework
>> draft in parallel to the registry design...
>
> +1
> I also see there is requirement from service provider, both on passive
measurement methodology and metrics. And I also see there's people working
on it in IPPM.
> I actually had a talk with Brian in London and Brian also see the
importance of a passive measurement framework. I'm personally interested in
working on this, and plan to submit the 00 framework draft before Toronto
meeting. People who has interest please contact me offline.
>
> BR,
> Vero
>
>> -----Original Message-----
>> From: ippm [mailto:[e-mail]] On Behalf Of Joachim Fabini
>> Sent: Wednesday, March 05, 2014 3:52 PM
>> To: Brian Trammell; [e-mail]
>> Subject: Re: [ippm] Fwd: Call for adoption: IPPM Registry Drafts
>>
>> Brian, IPPM,
>>
>> being one of the off-liners I support the call for adoption.
>> Two comments:
>>
>> 1. I DO support a metric registry which goes beyond the scope of IPPM.
>> There should be one central-point-of-contact where users and other
>> organizations can search for registries. Imho central topics with this
>> respect:
>> (a) to find out if the IETF process allows for this broad registry scope
as
>> mentioned during the meeting for the case of MPLS. I'm sure that there
are
>> other groups, too, defining metrics which fit into the registry. For
instance SIP
>> signaling metrics (RFC6076) might be another candidate, BMWG has, for
sure,
>> to add some more...
>> (b) to conduct a survey on which WGs would be affected, if/how their
>> requirements impact on the registry structure and how this input could be
>> gathered and fed into the registry design process.
>>
>> 2. Registries for active and passive metrics should be aligned to each
other, i.e.,
>> share as much as possible of structure and names. For instance consider
the
>> Packet-Type-P column for active measurements and the Filter Criteria
column
>> for passive measurements. These two imho might denote the same concept
>> with Packet-Type-P being formally defined in RFC2330. However, filtering
>> considers also packet content (runtime
>> information) as opposed to Packet-Type-P, which (iirc) affects only
packet
>> protocol/type/control information.
>> It might be also interesting to discuss to which extent sampling for
passive
>> measurements and traffic generation for active measurements use/share
>> identical methodologies and algorithms. A framework document for passive
>> measurements would help tremendously.
>>
>> Summarizing: the broader the scope of the registry documents is, the
higher
>> the registry value to (IETF-internal and -external) users will be.
Naming and
>> structure of active and passive registries should be aligned if feasible,
>> commonalities worked out.
>> Having no passive measurements framework document yet, my feeling is that
>> the proposed time schedule is pretty tight. The content of a
>> (future) passive measurement RFC will likely influence on the registry
structure.
>> I might be desirable to start work on the passive measurement framework
>> draft in parallel to the registry design...
>>
>> jm2c,
>> best regards
>> Joachim
>>
>> On 04.03.2014 17:07, Brian Trammell wrote:
>> > Greetings, all,
>> >
>> > At yesterday's IPPM WG meeting in London, we started a call for
adoption on
>> the following three drafts as working group items in support of our
charter item
>> to define a performance metric registry.
>> >
>> > draft-manyfolks-ippm-metric-registry-00
>> >
>> > draft-mornuley-ippm-registry-active-00
>> >
>> > draft-akhter-ippm-registry-passive-01
>> >
>> > There was clear consensus in the room to adopt, with the understanding
that
>> possible questions about the scope of the metrics defined in this
registry would
>> need to be addressed in the working group and across areas.
>> >
>> > I would propose the following milestones:
>> >
>> > Aug 2014 Submit draft on core registry for performance metrics to IESG
>> > as Proposed Standard Aug 2014 Submit draft on registry for active
>> > performance metrics to IESG as Proposed Standard Dec 2014 Submit draft
>> > on registry for passive performance metrics to IESG as Proposed
>> > Standard
>> >
>> > This would give us about six months (and one meeting) to discuss core
and
>> active registry issues, and nine months (two meetings) for passive
issues.
>> >
>> > We'd like to hear any comments on adoption from those not in the room,
>> and/or additional comments from all (especially on milestones and
timing) until
>> next Tuesday 11 March 2014.
>> >
>> > Thanks, best regards,
>> >
>> > Brian
>> >
>> > _______________________________________________
>> > ippm mailing list
>

-- 
邓灵莉/Lingli Deng
中国移动通信研究院/China Mobile Research Institute
e-mail: [e-mail]
tel: [phone]
mobile: [phone]