Glossary service vs. resource (RE: [ogsa-wg] Proposed agenda for Nov. 17th call)

[Hiro Kishimoto]

Hi all,

The attached is the agreed definition for service, resource, entity and 
manageable resource based on today's protracted discussion.

Fred's text (meat) is also included.

I will do global search glossary doc and ogsa v1 doc for possible
inconsistencies. I guess we don't have many.

Thanks,
----
Hiro Kishimoto


> -----Original Message-----
> From: owner-ogsa-wg at ggf.org [mailto:owner-ogsa-wg at ggf.org] On Behalf Of Hiro
> Kishimoto
> Sent: Thursday, November 18, 2004 7:28 AM
> To: 'ogsa-wg'
> Subject: Glossary service vs. resource (RE: [ogsa-wg] Proposed agenda for Nov.
17th
> call)
> 
> Hi all,
> 
> > From the telecon minutes of 20 Oct. 2004
> > Action: Hiro to do a first try on 'service' & 'resource' wrt the above
> discussion
> > and discuss with a small group before bringing back to the call.
> 
> The attached is my draft proposal for definition of service, resource, entity,
> manageable resource, and manageable entity with feedbacks form Dave
> Snelling and Fred Maciel.
> 
> My proposal is
> 
> (a) Service = Web service
> 
> Dave: This is OK, although some may argue that there are services of
>          interest that are not Web Services.
> 
> Fred: Agreed. We could mention that OGSA concerns only Web services,
>         and any service that's not a Web service is not part of the OGSA
>         framework.
> 
> Hiro: Agreed that OGSA v1.0 chooses Web service as its infrastructure.
>        But OGSA does not prohibit the other plumbing.
> 
> (b) Resource = entity (= stateful resource defined by WSRF)
> 
> Dave: WSRF-Resource(*) defines a resource is an entity that has
>          the following characteristics:
>          - It MUST be identifiable; a resource has at least one resource
>             identifier (see Section2.2).
>          - It MUST have a set of zero or more properties which are
>             expressible in XML infoset.
>          - It MAY have lifecycle.
> 
> Fred: I disagree. I think that in the real world not everything in a Grid will
>         be manageable. Suppose that you have a host that's not manageable
>         (say, you lost the root password -- you can only run jobs on it, press
>         the reset button if things go wrong, and nothing else). It is still an
> entity,
>         but I don't think that it is a resource. I think that the word
resource
>         implies that you may not only use it but also manipulate it (monitor
it,
> 
>         provision it or provision stuff on it, change its IP address or reboot
> it
>         remotely, etc.).
> 
> Hiro: Everything is "resource" but not "manageable resource."
> 
> (c) Manageable resource = manageable entity (will be represented by
WS-resource)
> 
> Dave: WSRF-Resource(*) defines a WS-Resource is a Web service
> 
> Hiro: It is tricky but OGSA does not define WS-Resource.
> 
> * the URL for WSRF-Resource:
> http://www.oasis-open.org/committees/download.php/9547/wsrf-WS-
> Resource-1.2-draft-01.doc
> 
> Thanks,
> ----
> Hiro Kishimoto

-------------- next part --------------
A non-text attachment was scrubbed...
Name: Service-resource v3.doc
Type: application/msword
Size: 41984 bytes
Desc: not available
Url : http://www.ogf.org/pipermail/ogsa-wg/attachments/20041118/4393a548/attachment.doc