Re: [Aeon] AECON Bar BOF at IETF90

["Fan, Peng" <[e-mail]>]

Hi all,

 

Please find the following minutes of the BarBOF, feel free to comment if
there is anything missing or incorrect. Thank you all for participating.

 

AECON barBoF

 

Logistics: Wednesday, July 23, 11:45-12:45, in Manitoba 

Attendance: About 30 people 

Overview: 

*	Information shared among the community about AECON, the progress so
far, meeting with IAB/IESG members, and feedback from that meeting
*	Open discussion about concerns of AECON and suggestions regarding
next steps

Notes taken: 

*	Security aspects, what is needed depends on what information is
communicated and the environment within which it is communicated; however,
there is general agreement we need to be more concrete about how AECON
addresses security 

*	Michael mentioned similar BoF failed because of security, need to
bear that in mind

*	It is clear that AECON could be misused by an operator; need to
illustrate how AECON makes things better 

*	Work out all the details for at least one use case, a prototype
would be helpful

*	Beware of privacy implications, even in home network 

*	Dad should not be able to see what kids are watching or what apps
they are using

*	Consider ALTO to home/first hop/edge router and what could be
achieved simply with AECON from application to home/first hop/edge router 

*	Quite a few people (10+) are familiar with ALTO
*	Need to work on gap analysis of ALTO, consider if it is possible to
extend ALTO to solve this

*	Need to hear from application developers how AECON helps 

*	Ease of implementation, lowering the bar to usage, and incremental
deployment are all targeted by AECON; need to engage more with application
developers to validate 
*	AECON targeted Concrete examples of what we want to do (prototype),
illustrate advantages and use cases to developers
*	Tim Chown reminded of a similar northbound API to SDN controller,
see how it can be used to describe AECON; need some experimental data, game
devs considered that
*	Pal-Erik to work on prototype to demonstrate at or before next IETF

*	Possibility of carrying characteristics within data plane 

*	It could be out-of-band or in-band, up to our discussion, one
concern about in-band solution is traffic encryption

*	Discussion about signaling with changing traffic, probably AECON
would need to send messages periodically and in response to specific events,
but the rate of such updates needs to be considered

*	one option discussed is having application initiate all AECON
signaling (e.g. network feedback in response to application signaling only),
other options exist

*	Explore what you can do with/without signaling an application
identifier 

*	Revisit existing use cases with this in mind

*	Revisit discussion from DiffServ WG 

*	Look at archives, reach out to chairs

*	Site page is out of date, is it useful, should we update it and
continue to use it? 

*	Yes, but definitely need to update in order for it to be helpful,
right now it is confusing

Best regards,

Peng

 

From: Aeon [mailto:[e-mail]] On Behalf Of Charles Eckel
(eckelcu)
Sent: Friday, July 25, 2014 6:34 AM
To: Fan, Peng; [e-mail]
Subject: Re: [Aeon] AECON Bar BOF at IETF90

 

For your reference, here are the slides from the barBOF. Comments and
questions are welcome.

We are still working on minutes. We hope to share them by tomorrow.

 

Cheers,

Charles

 

From: <Fan>, Peng <[e-mail]>
Date: Tuesday, July 22, 2014 at 10:33 AM
To: "[e-mail]" <[e-mail]>
Subject: Re: [Aeon] AECON Bar BOF at IETF90

 

Hi all,

 

The Bar BoF will be held in room Manitoba on the MM floor. The time is
Wednesday 11:45-12:45 Toronto local time.

 

Looking forward to seeing you tomorrow.

 

Best regards,

Peng

 

From: Aeon [mailto:[e-mail]] On Behalf Of Fan, Peng
Sent: Monday, July 21, 2014 10:03 AM
To: [e-mail]
Subject: [Aeon] AECON Bar BOF at IETF90

 

Hi all,

 

We will be having the AECON Bar BOF Wednesday (23 Jul) noon during
11:45-12:45. Meeting place will be notified soon. You are all welcome to
attend the discussion.

 

Best regards,

Peng