[occi-wg] XSD rendering of OCCI model(s)

Wed Dec 9 18:43:18 CST 2009

Many thanks for this useful contribution Costas! :-)

Andy

-----Original Message-----
From: occi-wg-bounces at ogf.org [mailto:occi-wg-bounces at ogf.org] On Behalf Of Constantinos (Costas) Kotsokalis
Sent: Tuesday, December 08, 2009 4:26 PM
To: Gary Mazz
Cc: occi-wg at ogf.org
Subject: Re: [occi-wg] XSD rendering of OCCI model(s)

Hi Gary,

Thanks for your reply. As regards the schema, I have attached to this email some first effort on it. Apologies if emailing attachments is not acceptable here, I couldn't find my way around uploading to the gridforge. There are two files attached, a schema that corresponds to the spec, and one that imports it to reuse some data types, and extends to a number of elements that (I think) would be more convenient for use w/ WS-Agreement. The main problem is that, in WS- Agreement, the links are not really necessary; rather, one would just declare a service description (e.g. "compute"), its properties in a separate place (e.g. cores, speed, memory) and then in a third place (s)he would set numbers for those properties (e.g. 2 cores, 1.5GHz, etc). The same would then happen for storage, and for network.  
Grouping is then handled by the WS-Agreement envelope, using "Term Compositors". Arbitrary possible combinations of different resource requirements would be a problem to express without this (seemingly
redundant) 2nd rendering in the additional schema file.

The schemas are still draft (read: a quick hack), so there may be many/ big mistakes. Please let me know if you have comments.

With regard to the cores data type, I am only puzzled by the
(syntactical) correctness of the expression "occi.compute.cores = 3.14159265" :) Nevertheless, the group has apparently had its discussions on the topic already, so it is fine.

Very best,

  Costas

On 8 Dec 2009, at 15:24, Gary Mazz wrote:

> Costas,
>
> I'm working on  a mapping of occi attributes to xds data types as part 
> of rdfa support in XHTML. We should coordinate efforts where
> possible. I'm tied up for most of this week, I'll make better   
> progress  over the weekend,,,,, it looks like I'll be snowed or frozen 
> in.
>
> Numerating data types is always an interesting topic on this  
> list :-)   This document is in the OGF pipeline for comment, if  
> there are sections of the document that are unclear, missing, 
> incompatibilities with other efforts, we'll be glad to address them.
>
> Vlans are a network artifact/feature supported by a few providers.  
> Consensus was reached there was sufficient provider support to warrant 
> an attribute in the networking resource.
> cheers,
> gary
>
>
> Constantinos (Costas) Kotsokalis wrote:
>> Hi Thijs,
>>
>> Not a problem. I have actually already created something but it's   
>> fairly simplistic as it doesn't capture links, and it also wouldn't  
>> help with WS-Agreement's concept of "service guarantees"
>> on "service  properties" of "service descriptions". So I'll update, 
>> try to take  into account the concepts from the existing schema that 
>> are still  valid according to latest spec, and send to the list.
>>
>> BTW, two question regarding the spec -- apologies in advance if these  
>> have been discussed before, as I just subscribed to the list. If so,  
>> then please just point me to the archived thread if possible.
>>
>> 1. occi.compute.cores -- why is this a float, instead of an integer?
>> 2. Is there a use case for occi.network.vlan? I'm slightly confused 
>> by  its use in a cloud context.
>>
>> Thanks,
>>
>>  Costas
>>
>>
>> On 8 Dec 2009, at 14:41, Thijs Metsch wrote:
>>
>>
>>> Hi Costas,
>>>
>>> Right now these are not synced...sorry - Probably time we a) update  
>>> it or b) remove it...:-) Can or do you wanna help with a)?
>>>
>>> Cheers,
>>>
>>> -Thijs
>>>
>>> On Dec 8, 2009, at 2:19 PM, Constantinos (Costas) Kotsokalis wrote:
>>>
>>>
>>>> Dear All,
>>>>
>>>> I would like to have an XSD rendering of the OCCI model(s), to 
>>>> embed in WS-Agreement [1]. I have found some previous work in [2], 
>>>> but I am not sure if this is synchronized with the latest spec 
>>>> draft(s) [3].
>>>> Could someone please confirm?
>>>>
>>>> Thank you in advance,
>>>>
>>>> Costas
>>>>
>>>> [1] http://www.ogf.org/documents/GFD.107.pdf
>>>> [2] 
>>>> http://forge.ogf.org/sf/wiki/do/viewAttachment/projects.occi-wg/wik
>>>> i/NounsVerbsAndAttributes/occi.xsd
>>>> [3] http://www.ogf.org/pipermail/occi-wg/2009-November/001517.html
>>>>
>>>> _______________________________________________
>>>> occi-wg mailing list
>>>> occi-wg at ogf.org
>>>> http://www.ogf.org/mailman/listinfo/occi-wg
>>>>
>>> -- 
>>> Thijs Metsch                        Tel: +49 (0)941 3075-122  
>>> (x60122)
>>> http://blogs.sun.com/intheclouds
>>> http://www.twitter.com/befreax
>>> Software Engineer Cloud, Grid and Virtualization Sun Microsystems 
>>> GmbH
>>> Dr.-Leo-Ritter-Str. 7                    mailto:thijs.metsch at sun.com
>>> D-93049 Regensburg                http://www.sun.com
>>>
>>>
>>
>> _______________________________________________
>> occi-wg mailing list
>> occi-wg at ogf.org
>> http://www.ogf.org/mailman/listinfo/occi-wg
>>
>>
>

-------------------------------------------------------------
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.