[DRMAA-WG] Conference call - November 23th - 19:00 UTC

[Peter Tröger]

Dear all,

since the public comment period for DRMAAv2 is now over, we need to
discuss our reaction on the comments. The next DRMAA conf call is
therefore scheduled for November 23th, 19:00 UTC.

We meet on Skype, please find me under the nickname "potsdam_pit".

Preliminary meeting agenda:

1. Meeting secretary for this meeting?
2. Public comments on DRMAAv2
(http://ogf.org/gf/docs/comment.php?id=377)
3. New co-chair election (details in the call)
4. Next steps with the C-binding

Best regards,
Peter.

[Peter Tröger]

Dear all,

I got numerous requests from key group members to shift this conf call
by one week. The new slot is therefore now November 30th, 19:00 UTC.

Best regards,
Peter.

Post by Peter Tröger
Dear all,
since the public comment period for DRMAAv2 is now over, we need to
discuss our reaction on the comments. The next DRMAA conf call is
therefore scheduled for November 23th, 19:00 UTC.
We meet on Skype, please find me under the nickname "potsdam_pit".
1. Meeting secretary for this meeting?
2. Public comments on DRMAAv2
(http://ogf.org/gf/docs/comment.php?id=377)
3. New co-chair election (details in the call)
4. Next steps with the C-binding
Best regards,
Peter.
--
drmaa-wg mailing list
drmaa-wg at ogf.org
http://www.ogf.org/mailman/listinfo/drmaa-wg

[Peter Tröger]

Participants: Mariusz, Daniel, Roger, Peter

Post by Peter Tröger
1. Meeting secretary for this meeting?

Peter

Post by Peter Tröger
2. Public comments on DRMAAv2
(http://ogf.org/gf/docs/comment.php?id=377)

-> "slots" attribute in SlotInfo

- Just a typo, proposed solution approved.

-> Attribute naming

- Agreement to stick with self-descriptive attribute names
- Agreement to current shortening approach in C binding draft:
- drmaa2_reservation_get_reservation_id (bad duplication)
- drmaa2_r_get_reservation_id (we like this approach)
- drmaa2_r_get_id (not self-descriptive enough)

-> exception extensibility

- Extensibility of exception list in language binding or implementation
- First one is easy (e.g. special expression of runtime issues in JVM)
- Does not break portability
- Allowing extensibility in implementations can break portability
- normally, we have introspection support for this
- Fetching the list of implementation-specific exceptions is useless
- Customer perspective: I want portability
- Vendor perspective: I want tailoring for special customer needs
- The Java classic: New exceptions must inherit from a DRMAA exception
- Solution: Introduction of a new ImplementationSpecificException
- Implementation-specific exception must inherit from this one
- Allows generic catch in portable application code
- Message parameter and numeric error id parameter
- Maps to custom error codes in C (no portability problem)

-> No CPU speed support
- not present in Torque, PBS, SGE; only in LSF with JSDL support
- intentionally left out, use job categories for that

-> Job / reservation closing on session closing
- Original intention was parent / child relationship
- Would lead to invalidation of all instances on session destruction
- Counter argument: You destroy uselessly fetched information
- Agreement: State that "on session close, job / reservation objects
MAY remain usable completely or in parts."

Post by Peter Tröger
3. New co-chair election (details in the call)

- Mariusz was elected as new co-chair, replacing Dan Templeton.

The group wants to express its gratitude for years of continuous support
and personal engagement by Dan. We are aware of the fact that Dan
(together with many other Sun employees) acted as the major driving
force behind the market acceptance of DRMAA. Hopefully, he will continue
to "just talk a lot" as part of the regular group work. And the world
anyway may want to see a DRMAA Hadoop binding ...

Post by Peter Tröger
4. Next steps with the C-binding

Still pending.

The next conf call will take place on Wednesday, 19:00 UTC.
(which means 20:00 for you, Peter).

Best regards,
Peter.

[Peter Tröger]

Dear all,

since the public comment period for DRMAAv2 C-binding is over, we need to
discuss our reaction on the comments. The next DRMAA conf call is
therefore scheduled for Wednesday, August 22, 19:00 UTC - 20:00 UTC.

We meet on Skype, please find me under the nickname "potsdam_pit".

Preliminary meeting agenda:

1. Meeting secretary for this meeting?
2. Status report by the participants
3. Public comments on DRMAAv2 C-binding
(http://ogf.org/gf/docs/comment.php?id=388)
- pointer NULLing by the library ? What it means for the interface ?
- Problems with OStype and CPUtype enumerations
- Semantics of DRMAA2_DATA_SEG_SIZE
- Additional free functions needed
- Return type for wait functions
- _j_ vs. _job_
- Small issues


Best regards,
Peter.


-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/pkcs7-signature
Size: 4343 bytes
Desc: not available
URL: <http://www.ogf.org/pipermail/drmaa-wg/attachments/20120820/999dbb12/attachment.bin>

[Andre Merzky]

Hi Peter,

I probably won't make the call, thus some input from my end. I don't
have overly strong opinions on any of the issues, so whatever option
gets the majority will be fine with me.

Comments inlined below.

Best, Andre.

Post by Peter Tröger
Dear all,
since the public comment period for DRMAAv2 C-binding is over, we need to
discuss our reaction on the comments. The next DRMAA conf call is
therefore scheduled for Wednesday, August 22, 19:00 UTC - 20:00 UTC.
We meet on Skype, please find me under the nickname "potsdam_pit".
1. Meeting secretary for this meeting?
2. Status report by the participants
3. Public comments on DRMAAv2 C-binding
(http://ogf.org/gf/docs/comment.php?id=388)
- pointer NULLing by the library ? What it means for the interface ?

I find nulling convenient and safe, and don't mind a slight asymmetry
in the interface to support it. You will find best practices in
support of it, but also good arguments against it, I am sure ;-)

Post by Peter Tröger
- Problems with OStype and CPUtype enumerations
- Semantics of DRMAA2_DATA_SEG_SIZE

Clarifying the text should suffice?

Post by Peter Tröger
- Additional free functions needed
- Return type for wait functions

return an error. Chaining is not very useful, and confusing, if not
applied to all calls - which would make error reporting difficult.

Post by Peter Tröger
- _j_ vs. _job_

_j_, as that will be frequently used. Not sure if ultra short
prefixes are good in general though. length versus readability is
hard to decide, and someone *will* be unhappy, no matter what you
choose...


Cheers, Andre.

Post by Peter Tröger
- Small issues
Best regards,
Peter.
--
drmaa-wg mailing list
drmaa-wg at ogf.org
https://www.ogf.org/mailman/listinfo/drmaa-wg

[Peter Tröger]

Post by Peter Tröger
Dear all,
since the public comment period for DRMAAv2 C-binding is over, we need to
discuss our reaction on the comments. The next DRMAA conf call is
therefore scheduled for Wednesday, August 22, 19:00 UTC - 20:00 UTC.
1. Meeting secretary for this meeting?

Participants: Andre, Daniel, Mariusz, Roger, Stephan Klauck, Peter
Meeting secretary: Peter

Post by Peter Tröger
2. Status report by the participants

All in good health, some reporting from Peter about OGF35 in Delft

Post by Peter Tröger
3. Public comments on DRMAAv2 C-binding
(http://ogf.org/gf/docs/comment.php?id=388)
- pointer NULLing by the library ? What it means for the interface ?

- No decision at OGF35, voting needed
- Roger, Peter: "principle of least astonishment, no NULLing in the library"
- Andre, Daniel, Mariusz: "security problem, asynchronous free problem, OpenMPI example"
- 3:2 voting result for changing the API -> free* functions get pointer variable reference and assign NULL to it
- Voting also clarified that no other API changes are made

Post by Peter Tröger
- Problems with OStype and CPUtype enumerations

- No decision at OGF35, voting needed
- 5:0 voting result for keeping MACOS identifier - OS version is already a separate concept in the API
- 5:0 voting result for changing TRUE64 to TRU64
- 5:0 voting result for adding 64bit variations of all instruction set architectures
- These changes demand an errata for the DRMAAv2 root specification

Post by Peter Tröger
- Semantics of DRMAA2_DATA_SEG_SIZE

- No decision at OGF35, voting needed
- 5:0 voting result for the proposal by Rayson Ho: DATA_SEG_SIZE is renamed to DATA_SIZE
- Explanation must extended - this value refers to 'initialized data, uninitialized data and heap'
- These changes demand an errata for the DRMAAv2 root specification

Post by Peter Tröger
- Additional free functions needed

- No decision at OGF35, voting needed
- 5:0 voting result for adding them

Post by Peter Tröger
- Return type for wait functions

- OGF35 decision acknowledged by Roger
- IDL spec should have void return type for wait() functions
- Additionally, language bindings should be allowed to change the return type for chaining support
- These changes demand an errata for the DRMAAv2 root specification

Post by Peter Tröger
- _j_ vs. _job_

- 4:1 voting result: Keep it as it is now

Post by Peter Tröger
- Small issues

- jobCategories should be accessible outside the job session
- Daniel: Need scheduler connection for fetching the information, and then again in the reservation session
- Mariusz: Making a reservation without some job submission afterwards is very unlikely
- 5:0 voting for keeping it as it is now
- Native drmaa2_string type, instead of char* pointers
- OGF35 decision acknowledged by Roger
- UNSET for JobInfo: 5:0 voting for adding it
- Remove endIndex upper limit specification demand from the root spec
- Demands an errata for the DRMAAv2 root specification

I will now start to prepare both the errata document and the final C binding for the editors.

Best regards,
Peter.


-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/pkcs7-signature
Size: 4343 bytes
Desc: not available
URL: <http://www.ogf.org/pipermail/drmaa-wg/attachments/20120828/5882e8f1/attachment.bin>

[Peter Tröger]

Dear all,

this is the final version of the DRMAAv2 C binding. Please check if your favorite last change was correctly incorporated. The document will go to the OGF editors in 24 hours.

Best regards,
Peter.

-------------- next part --------------
A non-text attachment was scrubbed...
Name: drmaa-c.pdf
Type: application/pdf
Size: 208970 bytes
Desc: not available
URL: <http://www.ogf.org/pipermail/drmaa-wg/attachments/20120903/214e80e7/attachment-0001.pdf>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: drmaa2.h
Type: application/octet-stream
Size: 18662 bytes
Desc: not available
URL: <http://www.ogf.org/pipermail/drmaa-wg/attachments/20120903/214e80e7/attachment-0001.obj>
-------------- next part --------------

Von: Peter Tr?ger <peter at troeger.eu>
Betreff: Meeting minutes - conference call - August 22
Datum: 28. August 2012 13:38:20 MESZ
An: "drmaa-wg at ogf.org" <drmaa-wg at ogf.org>

Post by Peter Tröger
Dear all,
since the public comment period for DRMAAv2 C-binding is over, we need to
discuss our reaction on the comments. The next DRMAA conf call is
therefore scheduled for Wednesday, August 22, 19:00 UTC - 20:00 UTC.
1. Meeting secretary for this meeting?

Participants: Andre, Daniel, Mariusz, Roger, Stephan Klauck, Peter
Meeting secretary: Peter

Post by Peter Tröger
2. Status report by the participants

All in good health, some reporting from Peter about OGF35 in Delft

Post by Peter Tröger
3. Public comments on DRMAAv2 C-binding
(http://ogf.org/gf/docs/comment.php?id=388)
- pointer NULLing by the library ? What it means for the interface ?

- No decision at OGF35, voting needed
- Roger, Peter: "principle of least astonishment, no NULLing in the library"
- Andre, Daniel, Mariusz: "security problem, asynchronous free problem, OpenMPI example"
- 3:2 voting result for changing the API -> free* functions get pointer variable reference and assign NULL to it
- Voting also clarified that no other API changes are made

Post by Peter Tröger
- Problems with OStype and CPUtype enumerations

- No decision at OGF35, voting needed
- 5:0 voting result for keeping MACOS identifier - OS version is already a separate concept in the API
- 5:0 voting result for changing TRUE64 to TRU64
- 5:0 voting result for adding 64bit variations of all instruction set architectures
- These changes demand an errata for the DRMAAv2 root specification

Post by Peter Tröger
- Semantics of DRMAA2_DATA_SEG_SIZE

- No decision at OGF35, voting needed
- 5:0 voting result for the proposal by Rayson Ho: DATA_SEG_SIZE is renamed to DATA_SIZE
- Explanation must extended - this value refers to 'initialized data, uninitialized data and heap'
- These changes demand an errata for the DRMAAv2 root specification

Post by Peter Tröger
- Additional free functions needed

- No decision at OGF35, voting needed
- 5:0 voting result for adding them

Post by Peter Tröger
- Return type for wait functions

- OGF35 decision acknowledged by Roger
- IDL spec should have void return type for wait() functions
- Additionally, language bindings should be allowed to change the return type for chaining support
- These changes demand an errata for the DRMAAv2 root specification

Post by Peter Tröger
- _j_ vs. _job_

- 4:1 voting result: Keep it as it is now

Post by Peter Tröger
- Small issues

- jobCategories should be accessible outside the job session
- Daniel: Need scheduler connection for fetching the information, and then again in the reservation session
- Mariusz: Making a reservation without some job submission afterwards is very unlikely
- 5:0 voting for keeping it as it is now
- Native drmaa2_string type, instead of char* pointers
- OGF35 decision acknowledged by Roger
- UNSET for JobInfo: 5:0 voting for adding it
- Remove endIndex upper limit specification demand from the root spec
- Demands an errata for the DRMAAv2 root specification
I will now start to prepare both the errata document and the final C binding for the editors.
Best regards,
Peter.

-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/pkcs7-signature
Size: 4343 bytes
Desc: not available
URL: <http://www.ogf.org/pipermail/drmaa-wg/attachments/20120903/214e80e7/attachment-0001.bin>