Liaison statement
Performance Measurement Liaison Response to Broadband Forum

Date: October 15, 2013

To:
Christophe Alter, Technical Committee Chair, Broadband Forum
(christophe.alter@orange.com)

From:
Dan Romascanu, IETF Large-Scale Measurement of Broadband Performance WG Chair
(dromasca@avaya.com) Jason Weil, IETF Large-Scale Measurement of Broadband
Performance WG Chair (jason.weil@twcable.com) Brian Trammell, IETF IP
Performance Metrics WG Chair (trammell@tik.ee.ethz.ch) Bill Cerveny, IETF IP
Performance Metrics WG Chair (bill@wjcerveny.com) Ray Bellis, IETF Home
Networking WG Chair (ray.bellis@nominet.org.uk) Mark Townsley, IETF Home
Networking WG Chair (mark@townsley.net) Sarah Banks, IETF Benchmarking
Methodology WG Chair (sbanks@aerohive.com) Al Morton, IETF Benchmarking
Methodology WG Chair (acmorton@att.com)

CC:
David Sinicrope, IETF/Broadband Forum Liaison Manager
(david.sinicrope@ericsson.com) Ross Callon, IETF Internet Architecture Board
(rcallon@juniper.net) Jari Arkko, IETF Chair (jari.arkko@piuha.net) Robin
Mersh, Broadband Forum CEO (rmersh@broadband-forum.org) Gabrielle Bingham,
Broadband Forum Secretariat (gbingham@broadband-forum.org) Jason Walls,
Broadband Home Working Group Co-Chair (jason@qacafe.com) John Blackford,
Broadband Home Working Group Co-Chair (john.blackford@pace.com) Dave Thorne,
Broadband Forum E2E Architecture WG Chair (david.j.thorne@bt.com) Dave Allan,
Broadband Forum E2E Architecture WG Chair (david.i.allan@ericsson.com) Sven
Ooghe, Broadband Forum E2E Architecture WG Vice Chair
(sven.ooghe@alcatel-lucent.com) Peter Adams, Broadband Forum Operations and
Network Management WG Chair (peter.adams@adtran.com)

Thank-you for your liaisons listed below and keeping the IETF in mind while
developing work work on Broadband Forumv(BBF) WT-304 Broadband Access Service
Attributes and Performance Metrics.

BBF Liaisons to date:
Mar 2013 - https://datatracker.ietf.org/liaison/1243/  to IETF Chair and IESG
Dec 2012 - https://datatracker.ietf.org/liaison/1221/  to IPPM chairs and
Transport and Ops ADs Sep 2012 - https://datatracker.ietf.org/liaison/1185/  to
IETF Chair and IESG Aug 2012 - https://datatracker.ietf.org/liaison/1179/  to
Ops Area Directors

Thank you also for your patience.  While there has been significant interest in
large-scale measurement of Broadband performance, formal IETF working groups to
address major components of this topic had not been chartered until this
summer.  Now that the IPPM WG has been re-chartered (to consider measurement
methods appropriate for large-scale measurements) and the LMAP WG formed (to
consider the architectural framework and operational components), both groups
look forward to communicating with the BBF on this subject.

In addition to LMAP and IPPM, you may also want to consider BMWG and Homenet
for some of your WT-304 efforts.  Please see the links below for all of these
working groups’ charters, scope and work plans.

LMAP (Ops and Mgmt Area) - https://datatracker.ietf.org/wg/lmap/charter/
IPPM (Transport Area) - https://datatracker.ietf.org/wg/ippm/charter/
BMWG (Ops and Mgmt Area) - https://datatracker.ietf.org/wg/bmwg/charter/
HomeNet (Internet Area) - https://datatracker.ietf.org/wg/homenet/charter/

We note that you would like to reference IETF protocols and work to satisfy the
requirements of your architecture.  We believe this is a good division of work
and scope and would be happy to cooperate along these lines.  We encourage
specific communication with each of the individual working groups via their
Chairs along the lines of their charters.  For topics that cross one or more
WGs, please address the liaison to the Chairs of all of the relevant working
groups and the liaison manager from the IETF who will help direct the liaison
to the appropriate WG or IETF authority.

The LMAP WG is reviewing your March 2013 Liaison and notes the following
requests for information: Feedback on the use of SIP to provide an inter and
intra domain mechanism to probe test target resource availability. Comments on
the use of DNS-SD(RFC6763) and mDNS (RFC6762) to support BBF's service
attribute communication. Information model development for test and report
parameters

LMAP WG Response:

The LMAP WG was formed in June and held its first meeting in July of this year.
Following the goals and milestones as outlined in the WG Charter that can be
found in the above link, the WG will be focussed on finalizing Use Case and
Framework documents and then beginning work on the Information Model document.
The Information Model was mentioned as an area of interest by the BBF in the
March Liaison. The LMAP WG would welcome input from the BBF as it begins work
on the Information Model document. For reference, the Informational Model scope
as described in the LMAP WG Charter is included for reference: Information
Model, the abstract definition of the information carried from the Controller
to the MA and the information carried from the MA to the Collector. It includes
   * The metric(s) that can be measured and values for its parameters such as
   the Peer MA participating in the measurement and the desired environmental
   conditions (for example, only conduct the measurement when there is no user
   traffic observed) * The schedule: when the measurement should be run and how
   the results should be reported (when and to which Collector) * The report:
   the metric(s) measured and when, the actual result, and supporting metadata
   such as location. Result reports may be organized in batches or may be
   reported immediately, such as for an on-demand measurement.

In regards to interest of using DNS-SD and mDNS in support of service attribute
communication between devices within the home network, the LMAP WG would defer
this question and possible further analysis to work that is taking place in the
Homenet WG. It should be noted that currently the mDNS Protocol is constrained
to a single link based on its use of link-local multicast. If the BBF would be
interested in the use of mDNS and DNS-SD in a multi-segmented home network,
this would require new work to those protocols that is being discussed as part
of a new Working Group. Discussion of that WG charter language is currently
taking place.

In regards to the use of SIP as a inter-domain and/or intra-domain mechanism
for the discovery of test target (Measurement Peer in LMAP terminology)
availability, this point represents communication that would take place between
a Measurement Agent and a Measurement Peer. Communication requirements as part
of a test between a MA and its measurement target (Measurement Peer) is
currently not included as one of the work items per the LMAP charter but may be
a topic of discussion for future work.

IPPM WG Response:

Please be advised that the IPPM WG has considered updates to RFC 2680 and 2681
in their March 2013 rechartering, but these have been deferred until it is
clear what impact the effort to update the framework (2330) and the effort to
define a registry on the March 2013 charter will have on these updates. IPPM
expects that the result will be compatible with 2680 and 2681, and as such
developments based on these RFCs may continue as they are.

Many of the issues with 3148 raised in Question 1 of the December 2012 BBF
liaison may be addressed by draft-ietf-ippm-model-based-metrics-00, on which
work is progressing under their current charter; discussion of issues with bulk
capacity measurement not addressed therein is welcome on the ippm@ietf.org
mailing list.

With respect to the remaining outstanding questions, the IPPM WG will take them
as information that BBF is considering the use of 3393 and 6349 as well; IPPM
is not considering updates to these at this time, so they remain a stable basis
for further work, noting again in the latter case that
draft-ietf-ippm-model-based-metrics-00 aims to address issues in using TCP to
measure TCP bulk transfer capacity.

IPPM understands from Question 6 that 'moving up the stack', looking
specifically at VoIP, streaming video, and DNS resolution time, are of interest
to BBF. IPPM is not currently working on metrics in this area, but would
certainly consider contributions thereon under its current charter, and have
reviewed at least one individual draft on buffered streaming video performance
during the chartering discussions in March 2013.

As for the general line of these questions (especially 6), please note the
ongoing registry effort on the IPPM charter. There are three individual drafts:
draft-bagnulo-ippm-new-registry, draft-bagnulo-ippm-new-registry-independent,
and draft-claise-ippm-perf-metric-registry. The authors are currently working
on a unified approach to a performance metrics registry, with the intention
that the outcome be adopted for further development within the IPPM WG. This
registry will be populated with recommended metrics for LMAP use cases, which
would represent a consensus statement from IPPM on the metrics IPPM consider
useful in this area. Work in this area is ongoing, and we welcome commentary on
the ippm@ietf.org list once the unified registry document is published.

We noted that there might be interest in the identification of test reference
points.  This is currently on the IPPM WG charter and we now have an initial
working group document
(http://datatracker.ietf.org/doc/draft-ietf-ippm-lmap-path).

The IETF encourages those in the BBF who are interested in this IETF work, to
participate and help drive the work via the relevant IETF WG email lists noted
above and the IETF development processes.  While formal liaison communication
is necessary and beneficial, direct, active participation by interested parties
is very helpful and complementary to drive the deliverables needed between the
organizations.  Please note that access to all relevant IETF working groups,
email lists, documents and process is open so than any interested party may
participate and contribute.

Likewise, the IETF would be happy to provide input and feedback on BBF related
work.  We understand the BBF is a membership organization and may restrict
access to its relevant works in progress.  To facilitate cooperation with the
IETF, please liaise any work in progress you would like the IETF to consider,
review, comment on, collaborate on, etc., with the understanding that access to
the liaison and its attachments will be open and not be restricted or limited
to BBF membership.

Sincerely,
LMAP, IPPM, Homenet and BMWG Chairs