[cddlm] RE: Notes from GGF16

[Stephen M Pickles]

Dejan,

All four sets of slides (including Steve's) from the Interoperability
Fests BoF should appear in the GGF site under GGF16 Materials in the
next few days. I'll post the URL when I have it.

I've requested an email list and a gridforge project as well.
I propose to post the notes there, once I've consolidated 
the notes from the two notetakers.

Thanks for your interest.

Stephen

> -----Original Message-----
> From: Milojicic, Dejan S (HP Labs) [mailto:dejan.milojicic at hp.com] 
> Sent: 22 February 2006 12:13
> To: cddlm-wg at ggf.org
> Cc: Stephen M Pickles
> Subject: Notes from GGF16
> 
> Hi,
> 
> Below is the summary of notes from GGF16. Thanks to all note takers. I
> am missing the 2/15 interop notes, Stephen, can you post them to this
> group, please. Steve, can you please post your presentation from the
> same session.
> 
> Best regards,
> 
> Dejan.
> 
> 
> CDDLM-WG (and related) Sessions at GGF 16
> 
> 2/13, 3:30pm: CDDLM #1, Demos of reference implementations
> 2/13, 5:30pm: CDDLM #2, Relationship to other technologies
> 2/14, 5:30pm: CDDLM #3, Planning interoperability
> 2/15, 5:30pm: BoF Interoperability Fests, CDDLM's participation
> 2/17: OGSA F2F, Continued discussions on OGSA (EMS) & CDDLM 
> 
> Quick Update on our Status and Where We are Going
> 
> - Documents
>   - Two documents are public
>   - Two documents finished public comment period, we expect them to
> become public documents shortly
>   - One document entering public comment 
> - Reference implementations
>   - Four in place with more or less completeness 
> - Interoperability
>   - Planned for March-May 2006, but some aspects already accomplished!
> - GGF17 (GGF18?) mission accomplished
> 
> --------------------------------------------------------------
> ----------
> ---------------------------
> CDDLM WG session #1, Demos of reference implementations (about 30
> people), Note Taker Hiro 
> --------------------------------------------------------------
> ----------
> ---------------------------
> 
> Introduction by Dejan
> Session plans in this week
> 
> Quick update
> Two document are published, two went through PC and will be published
> soon, and one starts PC soon.
> 
> Jun presents NEC implementation.
> It is stored in sourceforge.net, use common test environment, use CDL
> test (still growing).
> Test tool is using JUNIT.
> Handling multiple CDL documents to test <cdl:import>
> 
> "CDL processor" and "component generator"
> they are based on tomcat and muse with GUI.
> Demonstrate deployment and start "footwear shop" application.
> 
> What does component generator consume and generate? 
> 
> Show Ayla's demonstration without vocal.
> 
> Destroy undeploy the stuff? They do not do such optimization, yet.
> 
> What is the main difference between this demo and previous demo in
> Boston?
> Demo in Boston is just show it works. Now her implementation can take
> care of most of CDL specs.
> Do you have any statistics for her coverage?
> No. but this demo covers many corner cases and becomes much reliable.
> 
> Steven's SF based demo
> Show "hello world" demo
> Use VMware to demonstrate deploying networked machine.
> Demonstrate deploying jboss as more realistic example
> Do you also use CDL type for this window? No
> 
> GLARE presentation
> Askalon's grid environment
> Based on Globus services
> Problem is hardcoded binding between activity and component
> Want to move to runtime binding
> GLARE activity description = CDL?
> Super-peer model is distributed
> 
> Do you read CDLLM document? Many of features are already covered by
> them.
> We've implemented the system one ore more year before and are now
> reading specs.
> Dejan says they are very glad if you could give feedback in 
> the future.
> 
> Wrap up
> Dejan recaps another CDDLM sessions in this week.
> Dejan emphasizes they have four independent implementations.
> Meeting adjourned at 4:00pm 
> 
> --------------------------------------------------------------
> ----------
> --------------------------
> CDDLM WG session #2, Relationship to other technologies (about 30
> people), Note Taker Jun Tatemura
> --------------------------------------------------------------
> ----------
> --------------------------
> (participation from DMTF CIM and OASIS WSDM groups)
> 
> 
> relationship of CDDLM and BES
> to WSDM
> WSDM 1.1 (Fred)
> - referring to WSRF standardized version
> - WSRF expected March
> - problem: waiting for WS-Addressing
> 
> treat manageability as a profile...
> 
> 
> CDDLM - CIM
> - deployment artifact
> - underlying attribute model
> - may be good time to start
> thinking of mapping
> - need experts from DMTF?
> 
> 
> future interoperability fest
> DMTF + CDDLM + WSDM?
> - getting various groups
> to integrate a mock-up
> implementation
> 
> 
> impact from EGA merger
> - use cases available?
> 
> JSDL
> discussion
> initial idea on homework
> 
> SDD
> - they are now refining
> requirements
> 
> GRAAP
> Hiro explained OGSA EMS discussion/
> definition of deployment 
> and provisioning (policy) 
> 
> --------------------------------------------------------------
> ----------
> ---------------------
> CDDLM WG session #3, Planning interoperability  (about 30 
> people), Note
> Taker Dejan Milojicic
> --------------------------------------------------------------
> ----------
> ---------------------
> 
> CDL
> - Really in good shape
> - Do not have tests for some areas (lists, expressions)
> - Jun will complete this some time in March
> - Get status report from others where they are (e.g. 
> expressions as well
> as generic test reports)
> - Dejan will get the format for the interop document
> - Steve will automate tests inclusion
> - Lazy references not covered here (will be addressed by component
> model)
> - Jun/Steve will work on this, at least HP & NEC ref impl. but
> Softricity and UFCG
> 
> Basic deployment API
> - Need a common test client, we will have at least 2, Steve & TBD
> - HOWEVER: everyone should write his own test clients to be general
> enough to run against other servers
> - How to enforce that side effects are captured (create files 
> and assert
> content)
> - Also blurred with the deployment model, we need test components
> - See deployment test plan (WSRF properties, create errors, testing
> APIs)
> 
> Component Model (& Complete Deployment APIs) 
> - What is tested
>   - lifecycle
>   - notifications (??)
>   - file upload
> - Everyone should try to have publicly accessible servers (with
> passwords) so that we can try doing this ahead of the GGF17
> - CRITICAL:  common test components (even if all  different
> implementations for each team)
>   - file operations
>   - assertions
> 
> 
> --------------------------------------------------------------
> ----------
> ---------------------
> OGSA F2F, excerpt from Andreas' notes: 
> --------------------------------------------------------------
> ----------
> ---------------------
> CDDLM joint discussion
>   A JSDL BLAST example was posted to the list prior to GGF16. Steve
>   Loughran introduced a CDL example 'corresponding' to the JSDL
>   one. This immediately raised a number of questions on the approach
>   taken:
>   - What should the correspondence or relation between a JSDL and CDL
>     document? One could encapsulate the other; or there could be a
>     reference relation.
>   - What is the added value of using CDL? There is not much point in
>     simply duplicating the functions in the JSDL document.
> 
>   It was agreed that the CDL document should not just reproduce the
>   JSDL document but it should instead provide a more detailed
>   description of what needs to be set up to make it possible to run
>   the job described by the JSDL document.
>   - In part this is drawing the line between the deployment and
>     provisioning.
>   - As a general approach it was also agreed out that we should aim
>     for a relation such that 'moving the line' between
>     deployment/provisioning should not affect the contents of JSDL and
>     CDL documents.
> 
>   (There was also a repeat of the Sunnyvale (Jan 2006)
>   discussion. That is whether configuration actions should be done
>   implicitly by the container vs making them explicit. Re-affirmed the
>   consensus that configurations actions should be made explicit.)
> 
>   How close is CDL to be a declarative vs procedural language? Agreed
>   that it can probably be used either way, but with a preference
>   towards a declarative use for OGSA purposes.
> 
>   Proposal to work out details in a smaller group with people from the
>   various groups (JSDL, BES, CDDLM) before bringing it back for an
>   architecture review. 
>   - Accepted in principle, but was put aside in favor of the following
>     action.
> 
>   ACTION: Steven Newhouse to write up a first draft of a few more
>   scenarios on the relationship between JSDL and CDL.
>   - To be included in the current EMS scenarios document. 
>   - To be delivered by mid-March
>   - Review on an OGSA call and continue refinement
> 
> ** Whiteboard list produced during the discussion
> 
>   - Duplication of JSDL in CDL?
>   - Directory/user creation
>   - What is the difference between 
>  [job / application / container] configuration]
>   - Where is line between JSDL and CDL and is it policy specific
>   - Job lifecycle including deployment and start/stop
> 
> 
>