[occi-wg] OCCI 1.1 for OpenNebula

florian.feldhaus at tu-dortmund.de florian.feldhaus at tu-dortmund.de
Sat May 14 14:19:25 CDT 2011


Hi,

I just set up a server with our OCCI implementation using the dummy
backend. Feel free to try it at
http://ls22.itmc.tu-dortmund.de:4567/


If you like to use a demo server using the OpenNebula backend, please
contact me and I will configure access for you.

Cheers,
Florian

Am 13.05.11 18:42 schrieb "Feldhaus, Florian" unter
<florian.feldhaus at tu-dortmund.de>:

>Hi,
>
>as a followup to the announcement, I'd like to give some details and like
>to announce a second (alpha) release I just uploaded.
>
>Our OCCI Implementation is written in Ruby and we tried to make it
>independent from the backend. There currently are two backends
>implemented. Once a very simple Dummy Backend, which allows testing of the
>OCCI compliance. The other backend is interfacing between the OCCI server
>and OpenNebula using the OpenNebula Client API.
>
>As far as we understood the OCCI specification, our implementation is
>compliant with the OCCI Core and OCCI HTTP Rendering documents and it
>implements all required resources from OCCI Infrastructure and most of the
>Mixins (e.g. the Infrastructure templates are not yet implemented). If
>testing the implementation, please consider my comments regarding the
>specification which I summed up in my mail "Cloud Plugfest summary".
>
>The OpenNebula backend implements a subset of the features offered by our
>implementation. Currently the following is working:
>- GET is working for entities and kinds/mixins. The OCCI Server does not
>yet check for updates of status or parameters in OpenNebula.
>- POST is working for creation of entities. Actions are not yet supported,
>but will be soon - after some more testing.
>- PUT is not yet tested with the backend
>- DELETE is working
>
>During startup, all resources known to OpenNebula are initialized by the
>OCCI server and can be used through OCCI requests. Images can be uploaded
>during creation of OCCI Storage resources. OpenNebula specific features
>which could not be mapped to OCCI parameters were implemented as
>OpenNebula specific Mixins.
>
>During the implementation we had a lot of problems mapping the OCCI
>Infrastructure model to the OpenNebula model. From our point of view it
>would be best to specify only a very small set of mandatory parameters for
>the Infrastructure resources and then use a model like CIM for more
>complex requests. The OCCI Infrastructure model seems to be the central
>point for interoperability and thus a lot of experience is required to
>find a suitable "best practise". Are there any documents / discussions I
>can read to get more information on the decision / next steps for the
>Infrastructure model? I would be glad to participate in finding best
>practices for interoperability.
>
>I will probably attend OGF 32, who else does?
>
>Also there was an announcement at the end of the last SNIA plugfest that
>another plugfest should take place just before OGF 32. If so, I would also
>very much like to attend in person. Does anyone have more information on
>an upcoming plugfest?
>
>Cheers,
>Florian
>
>Am 12.05.11 11:40 schrieb "Daniel Molina" unter <dmolina at opennebula.org>:
>
>>Hi,
>>
>>The OpenNebula Team is happy to announce a new contribution to the
>>ecosystem, OCCI 1.1 for OpenNebula.
>>For more information, please visit: http://blog.opennebula.org/?p=1596
>>
>>We would like to thank Florian Feldhaus and Piotr Kasprzak for this
>>new component.
>>
>>Regards
>>
>>-- 
>>Daniel Molina, Cloud Technology Engineer/Researcher
>>Major Contributor
>>OpenNebula - The Open Source Toolkit for Cloud Computing
>>www.OpenNebula.org | dmolina at opennebula.org
>>_______________________________________________
>>occi-wg mailing list
>>occi-wg at ogf.org
>>http://www.ogf.org/mailman/listinfo/occi-wg
>



More information about the occi-wg mailing list