[dcifp-bof] Group charter proposal

[Bernhard Schott]

Hey Thijs, Alexander,
I am in support of the term "protocol" rather than "API" as our aim is to stay agnostic to the actually used technology. No matter which Java or C or Lego-Mindstorm somebody wants to use, it should be possible to implement the DCI-Federation-Protocol.

I am expecting support by some projects (e.g. DGSI, ...) to implement the federation protocol with their respective choice of framework etc and potentially deliver APIs as a side effect.

Best regards

Bernhard


-----Ursprüngliche Nachricht-----
Von:	dcifp-bof-bounces at ogf.org im Auftrag von Alexander Papaspyrou
Gesendet:	Mo 31.05.2010 17:07
An:	Thijs Metsch
Cc:	dcifp-bof at ogf.org
Betreff:	Re: [dcifp-bof] Group charter proposal

Hey Thijs,

Am 31.05.2010 um 16:59 schrieb Thijs Metsch:

> Or option 3) remove references to Grid & cloud and just leave the
> description you already have in there:
> 
> The DCI Federation Protocol Working Group (DCIFP-WG) will deliver an API
> specification for the automatically negotiated, SLA-secured, dynamically
> provisioned federation of resources and. The scope of the specification
> will include (1) the delegation
> of workload from one domain into the other, covering description,
> submission,
> and monitoring; and (2) leasing of resources, including definition,
> provisioning, and monitoring.

good idea, +1 from me. Alas, we will have to explicitly state (at least) Grid in the "who will uptake" Q6, I think.

> Second question would be: API vs Protocol???

That's also not entirely clear to me and partly the reason for making the bold statement of "protocols and APIs" here and there.

I think that for the "automatically negotiated" part, we will have to specify some kind of protocol, even if it boils down to a rather simple sequence diagram. Also, I probably tend to use the term "protocol" loosely here: I certainly do not want to define a new Over-The-Wire protocol -- just thinking of "steps to take to get a delegation/lease done".

HTH.

Best,
Alexander

>> -----Original Message-----
>> From: dcifp-bof-bounces at ogf.org [mailto:dcifp-bof-bounces at ogf.org] On
>> Behalf Of Edmonds, AndrewX
>> Sent: Mittwoch, 26. Mai 2010 19:17
>> To: Alexander Papaspyrou; dcifp-bof at ogf.org
>> Subject: Re: [dcifp-bof] Group charter proposal
>> 
>> On a second read...
>> Another aspect to consider is whether the charter should to have a
> wide
>> scope (cloudcomputing && gridcomputing) or narrow/focused scope
>> (cloudcomputing || gridcomputing). My own (biased ;-)) opinion on this
>> would be that a narrow and focused scope is taken.
>> 
>> Andy
>> 
>> -----Original Message-----
>> From: dcifp-bof-bounces at ogf.org [mailto:dcifp-bof-bounces at ogf.org] On
>> Behalf Of Edmonds, AndrewX
>> Sent: Wednesday, May 26, 2010 1:44 PM
>> To: Alexander Papaspyrou; dcifp-bof at ogf.org
>> Subject: Re: [dcifp-bof] Group charter proposal
>> 
>> Hey Alexander,
>> Great start and the charter reads very well!
>> 
>> In the charter (Q1) it says that the WG will define a new API. Granted
>> that might be an eventual necessity and outcome/output but should it
>> also be captured in the charter that suggested
> extensions/modifications
>> to the listed existing standards be developed and provided as an
> output
>> of the group and input to other OGF WGs? What I'm suggesting here is
> to
>> avoid the recreation of existing APIs/protocols by using the new API
> as
>> a proxy to them.
>> 
>> What's described in the charter would be of interest to SLA at SOI and I
>> would imagine the SLA at SOI/RESERVOIR collaboration group that is
>> currently active (I've mailed out to that group) would have an
> interest
>> too.
>> 
>> Andy
>> 
>> -----Original Message-----
>> From: dcifp-bof-bounces at ogf.org [mailto:dcifp-bof-bounces at ogf.org] On
>> Behalf Of Alexander Papaspyrou
>> Sent: Wednesday, May 26, 2010 11:02 AM
>> To: dcifp-bof at ogf.org
>> Subject: [dcifp-bof] Group charter proposal
>> 
>> Dear DCIFP participants,
>> 
>> our stated goal for OGF30 was to form a "real" group, that is move
> away
>> from our current BoF status. For those not familiar with the OGF
>> process, this requires a formal group creation request to the Grid
>> Forum Steering Group (GFSG) and a charter describing the purpose,
>> goals, and scope of the group, and its position within OGF and other
>> standardization organizations.
>> 
>> I have written a draft charter (see below) on the basis of what I've
>> found from other successfully established groups and our discussions
> in
>> Munich.
>> 
>> Please review and comment!
>> 
>> Best,
>> Alexander
>> 
>> --8<-- snip --8<--
>> -------------------------------------------------------------
>> Intel Ireland Limited (Branch)
>> Collinstown Industrial Park, Leixlip, County Kildare, Ireland
>> Registered Number: E902934
>> 
>> This e-mail and any attachments may contain confidential material for
>> the sole use of the intended recipient(s). Any review or distribution
>> by others is strictly prohibited. If you are not the intended
>> recipient, please contact the sender and delete all copies.
>> 
>> _______________________________________________
>> dcifp-bof mailing list
>> dcifp-bof at ogf.org
>> http://www.ogf.org/mailman/listinfo/dcifp-bof
>> -------------------------------------------------------------
>> Intel Ireland Limited (Branch)
>> Collinstown Industrial Park, Leixlip, County Kildare, Ireland
>> Registered Number: E902934
>> 
>> This e-mail and any attachments may contain confidential material for
>> the sole use of the intended recipient(s). Any review or distribution
>> by others is strictly prohibited. If you are not the intended
>> recipient, please contact the sender and delete all copies.
>> 
>> _______________________________________________
>> dcifp-bof mailing list
>> dcifp-bof at ogf.org
>> http://www.ogf.org/mailman/listinfo/dcifp-bof
> 

-- 
Alexander Papaspyrou
alexander.papaspyrou at tu-dortmund.de



-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://www.ogf.org/pipermail/dcifp-bof/attachments/20100531/f37f79be/attachment.html