Thanks! @Sam: can you update us a bit on the CloudAudit/OCCI connection? I'm almost always unable to attend the CloudAudit calls :-( Thanks, -Thijs On Wed, 2010-03-03 at 08:32 -0700, Mark A. Carlson wrote:
From the ongoing calls.
-- mark _______________________________________________ occi-wg mailing list occi-wg@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@sun.com D-93049 Regensburg http://www.sun.com
Thijs, Good question. I was going to raise this today. This is something which is still not fully defined. We have the concept of a "glue code" which need to be elaborated. Maybe, a call just for the "glue code" might be in order. Cheers <k/> On 3/3/10 Wed Mar 3, 10, "Thijs Metsch" <Thijs.Metsch@Sun.COM> wrote:
Thanks!
@Sam: can you update us a bit on the CloudAudit/OCCI connection? I'm almost always unable to attend the CloudAudit calls :-(
Thanks,
-Thijs
On Wed, 2010-03-03 at 08:32 -0700, Mark A. Carlson wrote:
From the ongoing calls.
-- mark _______________________________________________ occi-wg mailing list occi-wg@ogf.org http://www.ogf.org/mailman/listinfo/occi-wg
Thijs/Krishna/OCCI-WG, So I haven't said much on the subject yet, beyond highlighting the opportunity, but I think it would be sensible for us to list CloudAudit (A6) alongside CDMI as a compatible/suggested/recommended API. Like CDMI, CloudAudit provides important functionality that is just outside of the OCCI charter - rather than bloat OCCI it makes a lot more sense for us to align ourselves with other groups. Thus far CloudAudit can coexist with OCCI at the same API entry point, so it is just a case of advertising its existence to OCCI clients and/or suggesting that implementors may want to implement both specs together. It does this by using an unusual but standardised directory: /.well-known/cloudaudit/ (the .well-known directory was created for services like robots.txt so as to avoid polluting the root namespace). We have done two things in the current proposals to foster adoption of CloudAudit - made the v1 API dead simple (it can thus far be implemented with 0 LOC, simply by creating files on a web server) and provided an optional resolution service at cloudaudit.net to make it immediately useful (you can play in our sandbox at http://cloudaudit.net/.well-known/cloudaudit/). It is likely that CloudAudit will be useful for OCCI implementors and vice versa - thus facilitating the adoption of both APIs. You would be surprised how flexible this approach is: - detection is trivial - implementation is fast & costs are negligible (contrast to months to get even a simple REST API into production at a large site) - like OCCI we can support any format: - simple boolean primitives (`touch <file>`) - structured formats (xml, json, etc.) - human-friendly formats (PDF printouts or TIFF faxes of auditor reports, DOC/ODF originals, etc.) - compatible with other APIs, websites, etc. co-located at the same URI (e.g. http://cloud.example.com) - all the usual caching/proxying goodness of HTTP - security (confidentiality, integrity by way of SSL, availability by way of DNS) Technically the integration would likely be implemented as a <link rel="http://cloudaudit.org/" href="http://cloud.example.com" title="CloudAudit" /> or similar. Anyway for more information or to join visit http://www.cloudaudit.org/ (and A6 people can find out more about OCCI at http://www.occi-wg.org/) Sam (who was in a data center today and thus unable to join the call(s)) On 3/3/10, Krishna Sankar <ksankar42@gmail.com> wrote:
Thijs, Good question. I was going to raise this today. This is something which is still not fully defined. We have the concept of a "glue code" which need to be elaborated. Maybe, a call just for the "glue code" might be in order. Cheers <k/>
On 3/3/10 Wed Mar 3, 10, "Thijs Metsch" <Thijs.Metsch@Sun.COM> wrote:
Thanks!
@Sam: can you update us a bit on the CloudAudit/OCCI connection? I'm almost always unable to attend the CloudAudit calls :-(
Thanks,
-Thijs
On Wed, 2010-03-03 at 08:32 -0700, Mark A. Carlson wrote:
From the ongoing calls.
-- mark _______________________________________________ occi-wg mailing list occi-wg@ogf.org http://www.ogf.org/mailman/listinfo/occi-wg
On rare occasions I do agree with Sam ;o) and this happens to be one of them! Sam, yep, good thought. CloudAudit can be a capability that works along side OCCI and CDMI. In fact, the OCCI/CDMI/CloudAudit syndication is easier than OCCI/CDMI, as in the latter we have to exchange metadata, control path information as well as figure out the data path. Cheers <k/> P.S : Am surprised that they let you in the datacenter ;o) On 3/3/10 Wed Mar 3, 10, "Sam Johnston" <samj@samj.net> wrote:
Thijs/Krishna/OCCI-WG,
So I haven't said much on the subject yet, beyond highlighting the opportunity, but I think it would be sensible for us to list CloudAudit (A6) alongside CDMI as a compatible/suggested/recommended API. Like CDMI, CloudAudit provides important functionality that is just outside of the OCCI charter - rather than bloat OCCI it makes a lot more sense for us to align ourselves with other groups.
Thus far CloudAudit can coexist with OCCI at the same API entry point, so it is just a case of advertising its existence to OCCI clients and/or suggesting that implementors may want to implement both specs together. It does this by using an unusual but standardised directory: /.well-known/cloudaudit/ (the .well-known directory was created for services like robots.txt so as to avoid polluting the root namespace).
We have done two things in the current proposals to foster adoption of CloudAudit - made the v1 API dead simple (it can thus far be implemented with 0 LOC, simply by creating files on a web server) and provided an optional resolution service at cloudaudit.net to make it immediately useful (you can play in our sandbox at http://cloudaudit.net/.well-known/cloudaudit/). It is likely that CloudAudit will be useful for OCCI implementors and vice versa - thus facilitating the adoption of both APIs.
You would be surprised how flexible this approach is: - detection is trivial - implementation is fast & costs are negligible (contrast to months to get even a simple REST API into production at a large site) - like OCCI we can support any format: - simple boolean primitives (`touch <file>`) - structured formats (xml, json, etc.) - human-friendly formats (PDF printouts or TIFF faxes of auditor reports, DOC/ODF originals, etc.) - compatible with other APIs, websites, etc. co-located at the same URI (e.g. http://cloud.example.com) - all the usual caching/proxying goodness of HTTP - security (confidentiality, integrity by way of SSL, availability by way of DNS)
Technically the integration would likely be implemented as a <link rel="http://cloudaudit.org/" href="http://cloud.example.com" title="CloudAudit" /> or similar.
Anyway for more information or to join visit http://www.cloudaudit.org/ (and A6 people can find out more about OCCI at http://www.occi-wg.org/)
Sam (who was in a data center today and thus unable to join the call(s))
On 3/3/10, Krishna Sankar <ksankar42@gmail.com> wrote:
Thijs, Good question. I was going to raise this today. This is something which is still not fully defined. We have the concept of a "glue code" which need to be elaborated. Maybe, a call just for the "glue code" might be in order. Cheers <k/>
On 3/3/10 Wed Mar 3, 10, "Thijs Metsch" <Thijs.Metsch@Sun.COM> wrote:
Thanks!
@Sam: can you update us a bit on the CloudAudit/OCCI connection? I'm almost always unable to attend the CloudAudit calls :-(
Thanks,
-Thijs
On Wed, 2010-03-03 at 08:32 -0700, Mark A. Carlson wrote:
From the ongoing calls.
-- mark _______________________________________________ occi-wg mailing list occi-wg@ogf.org http://www.ogf.org/mailman/listinfo/occi-wg
Hi, thanks for the update! Sounds very interesting and I totally agree with you on the usage of OCCI and CloudAudit in conjunction. Sounds cool - I'll try to free some time and join the CloudAudit calls from time to time. Have fun, -Thijs P.S. I bet the datacenter you visited had not one of those 'cute' sun/oracle logos on it :-) On Thu, 2010-03-04 at 03:29 +0100, Sam Johnston wrote:
Thijs/Krishna/OCCI-WG,
So I haven't said much on the subject yet, beyond highlighting the opportunity, but I think it would be sensible for us to list CloudAudit (A6) alongside CDMI as a compatible/suggested/recommended API. Like CDMI, CloudAudit provides important functionality that is just outside of the OCCI charter - rather than bloat OCCI it makes a lot more sense for us to align ourselves with other groups.
Thus far CloudAudit can coexist with OCCI at the same API entry point, so it is just a case of advertising its existence to OCCI clients and/or suggesting that implementors may want to implement both specs together. It does this by using an unusual but standardised directory: /.well-known/cloudaudit/ (the .well-known directory was created for services like robots.txt so as to avoid polluting the root namespace).
We have done two things in the current proposals to foster adoption of CloudAudit - made the v1 API dead simple (it can thus far be implemented with 0 LOC, simply by creating files on a web server) and provided an optional resolution service at cloudaudit.net to make it immediately useful (you can play in our sandbox at http://cloudaudit.net/.well-known/cloudaudit/). It is likely that CloudAudit will be useful for OCCI implementors and vice versa - thus facilitating the adoption of both APIs.
You would be surprised how flexible this approach is: - detection is trivial - implementation is fast & costs are negligible (contrast to months to get even a simple REST API into production at a large site) - like OCCI we can support any format: - simple boolean primitives (`touch <file>`) - structured formats (xml, json, etc.) - human-friendly formats (PDF printouts or TIFF faxes of auditor reports, DOC/ODF originals, etc.) - compatible with other APIs, websites, etc. co-located at the same URI (e.g. http://cloud.example.com) - all the usual caching/proxying goodness of HTTP - security (confidentiality, integrity by way of SSL, availability by way of DNS)
Technically the integration would likely be implemented as a <link rel="http://cloudaudit.org/" href="http://cloud.example.com" title="CloudAudit" /> or similar.
Anyway for more information or to join visit http://www.cloudaudit.org/ (and A6 people can find out more about OCCI at http://www.occi-wg.org/)
Sam (who was in a data center today and thus unable to join the call(s))
On 3/3/10, Krishna Sankar <ksankar42@gmail.com> wrote:
Thijs, Good question. I was going to raise this today. This is something which is still not fully defined. We have the concept of a "glue code" which need to be elaborated. Maybe, a call just for the "glue code" might be in order. Cheers <k/>
On 3/3/10 Wed Mar 3, 10, "Thijs Metsch" <Thijs.Metsch@Sun.COM> wrote:
Thanks!
@Sam: can you update us a bit on the CloudAudit/OCCI connection? I'm almost always unable to attend the CloudAudit calls :-(
Thanks,
-Thijs
On Wed, 2010-03-03 at 08:32 -0700, Mark A. Carlson wrote:
From the ongoing calls.
-- mark _______________________________________________ occi-wg mailing list occi-wg@ogf.org http://www.ogf.org/mailman/listinfo/occi-wg
_______________________________________________ occi-wg mailing list occi-wg@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@sun.com D-93049 Regensburg http://www.sun.com
participants (4)
-
Krishna Sankar -
Mark A. Carlson -
Sam Johnston -
Thijs Metsch