Re: [clouds] IETF-78 Cloud Standards BOF

[Sam Johnston <[e-mail]>]

Bob/Monique/Linda,

I personally am very sceptical of (and would vigorously oppose) any attempt
to *coordinate* standards in the cloud space and echo Andy Edmonds in
pointing at cloud-standards.org (which already purports to do so). I am
however fully supportive of *collaboration* between standards orgs and
believe cloud-standards.org is as good a (neutral) venue as any, having
participated in a number of their calls. I also don't believe a group exists
(or even could exist) which would be able to objectively assess the cloud
space as each views the problem from a completely different perspective - in
this case the underlying telecommunications infrastructure while one could
argue cloud is more user-centric. Nevertheless the activity may prove useful
for ITU-T's internal use.

What was abundantly clear at today's bar bof is that there are a number of
areas where the IETF can deliver significant value now and we should focus
on these rather than participating in the landrush going on around us - as I
said in my presentation, there are currently no standards in cloud, only
specifications.

Getting back to Chris's original question, I don't think there's much point
in spending any time creating yet another model for cloud, however
enumerating specific potential work areas (e.g. CloudAudit, CloudLog, HTTP
enhancements, ARP222, etc.) would be time well spent... particularly if it
were to result in drafts as is the case for these examples.

Cheers,

Sam

On 28 July 2010 19:11, Natale, Bob <[e-mail]> wrote:

>  Hi Monique,
>
>
>
> I fully agree.  In fact, I meant my first sentence below to read "…of
> having a single SDO **coordinate** efforts…." … did **not** mean to imply
> that there should be only a single SDO in the cloud space, for sure!  Not
> meant to stifle individual SDO initiative, innovation, or momentum … just
> introduce some rational division of labor and coordination of
> effort/deliverables at the collective SDO level.
>
>
>
> Cheers,
>
> BobN
>
>
>
> *From:* Monique Morrow (mmorrow) [mailto:[e-mail]]
> *Sent:* Wednesday, July 28, 2010 1:03 PM
> *To:* Natale, Bob; Linda Dunbar; Chris Fenton (Iridescent);
> [e-mail]
>
> *Cc:* [e-mail]
> *Subject:* RE: [clouds] IETF-78 Cloud Standards BOF
>
>
>
> Bob,
>
> With ITU-T Focus Group Cloud Computing Vice Chair hat on, this suggestion
> is pragmatic.
>
> In the meantime, as IETF contributor, want to additionally focus on
> centering in on our use cases-scenarios- applicability of IETF-developed
> protocols and/or identification of protocols that should be developed within
> the IETF accordingly.
>
> With warm regards
>
> Monique
>
> PS BoF continues tomorrow!
>
> -----Original Message-----
> From: Natale, Bob [mailto:[e-mail] <[e-mail]>]
> Sent: Wed 7/28/2010 9:13 AM
> To: Monique Morrow (mmorrow); Linda Dunbar; Chris Fenton (Iridescent);
> [e-mail]
> Cc: [e-mail]
> Subject: RE: [clouds] IETF-78 Cloud Standards BOF
>
> Hi,
>
> I think Linda's idea of having a single SDO - and the ITU-T works fine for
> me in this case - is a good one.
>
> If the IETF endorses it, I could convey that message to the TM Forum Cloud
> Services Initiative as well.
>
> Each such SDO that agrees with the approach would send a liaison message to
> the ITU-T Cloud group expressing agreement, stating the focus area(s) best
> assigned to that SDO (e.g., the IETF-based protocols mentioned by Monique
> below), identifying any scheduled near-term (1 yr or less?) deliverables in
> those areas, and soliciting future "assignments" (to be negotiated via open
> collaboration) from the ITU-T.  With the stipulation that the ITU-T Cloud
> effort must be open in terms of meetings, documents, and inputs at a
> minimum.  Or something like that...?
>
> Cheers,
> BobN
>
> From: [e-mail] [mailto:[e-mail]<[e-mail]>]
> On Behalf Of Monique Morrow (mmorrow)
> Sent: Wednesday, July 28, 2010 10:37 AM
> To: Linda Dunbar; Chris Fenton (Iridescent); [e-mail]
> Cc: [e-mail]
> Subject: Re: [clouds] IETF-78 Cloud Standards BOF
>
>
> Linda
>
> That is certainly an approach -- in addition to leveraging and developing
> IETF-based protocols as discussed today in some of the use cases presented.
>
> Best -monique
>
>
>
> -----Original Message-----
> From: Linda Dunbar [mailto:[e-mail] <[e-mail]>]
> Sent: Wed 7/28/2010 6:20 AM
> To: Monique Morrow (mmorrow); 'Chris Fenton (Iridescent)'; [e-mail]
> Cc: [e-mail]; 'Bhumip Khasnabish'
> Subject: RE: [clouds] IETF-78 Cloud Standards BOF
>
> It is very nice that Bhumip to create a bar BOF for interested parties to
> present different aspects of Cloud.  As we can see that API and Cloud log
> are very different from dynamic resource management, which is also
> different
> from Private VPN. They need different expertise. People in Service layer
> may
> think the network wiring too detail and trivial. People worrying about
> network details view API as software features which can be application
> specific.
>
> Since ITU-T already starts the Cloud initiative, why not let ITU-T defines
> the general framework for Cloud and come to IETF for specific problems?
>
>
>
> Linda Dunbar
>
>
>
>   _____
>
> From: [e-mail] [mailto:[e-mail]<[e-mail]>]
> On Behalf Of
> Monique Morrow (mmorrow)
> Sent: Wednesday, July 28, 2010 7:31 AM
> To: Chris Fenton (Iridescent); [e-mail]
> Cc: [e-mail]
> Subject: Re: [clouds] IETF-78 Cloud Standards BOF
>
>
>
> Chris
>
> Do you proposed perhaps a framework draft as a starting point?
>
> Monique
>
>
>
>
>
> -----Original Message-----
> From: Chris Fenton (Iridescent) [mailto:[e-mail]<[e-mail]>
> ]
> Sent: Wed 7/28/2010 5:02 AM
> To: [e-mail]
> Cc: [e-mail]
> Subject: [clouds] IETF-78 Cloud Standards BOF
>
> Hi all,
>
> Thanks for interesting presentations.  To me the cloud without common or
> option for a standardised interface/api is not a cloud its a set of service
> platforms. We need to move forwards define the topic areas for
> standardisation but essentially its all the elements you have for
> subscribers to access a data platform from identity thru transport. Then
> there are the service management aspects and the service provider
> capabilities who want logs session or usage records and finally the
> platform
> owner would need some or at least be able to support the above....
>
> So possible topic areas for definition are :
>
> Service user
> - access
> - transport
> - service discovery maybe
>
> Service provider
> - Service management (also VM management - movement etc...)
> - Subscription management
> - charging billing etc...
> - basically these are all FCAPS aspects Fault, Configuration, Accounting,
> Performance, Security
>
> Platform provider - not so sure what standards apply so long as the
> platform
> itself provides the APis the boxes can be anything
> - Management
> - also FCAPS
>
> CJ
>
> Mr Chris Fenton
> [phone]
> [e-mail]
> Skype: chrisfentouk
> http://www.iridescentnetworks.com
>
>
>
>
>
>
>
> _______________________________________________
> clouds mailing list
> [e-mail]
> https://www.ietf.org/mailman/listinfo/clouds
>
>