
Our call schedule is roughly bi-weekly, so the next call would be in two weeks time (Mar.26). But it looks like the chairs won't be available that week. We are therefore thinking of a call next week, conditional on having sufficient material for discussion. We discussed parameter sweep yesterday and we probably won't have enough material for next week on this topic. That leaves the Activity schema, for which the main action from OGF22 was to provide use cases (Chris, Shabaz, Philipp). So could you please let me know if the use cases are likely to be ready by next week? I'll send out a call announcement if there is sufficient material, otherwise the next call will probably be in early April. -- Andreas Savva

Andreas, all, for the Grid Scheduling use case, I won't make it until next week. As such, I would generally participate *if* there is a call, but not protest if we decide to cancel it. Philipp, what is your opinion on this? -Alexander 2008/3/13, Andreas Savva <andreas.savva@jp.fujitsu.com>:
Our call schedule is roughly bi-weekly, so the next call would be in two weeks time (Mar.26). But it looks like the chairs won't be available that week.
We are therefore thinking of a call next week, conditional on having sufficient material for discussion. We discussed parameter sweep yesterday and we probably won't have enough material for next week on this topic. That leaves the Activity schema, for which the main action from OGF22 was to provide use cases (Chris, Shabaz, Philipp). So could you please let me know if the use cases are likely to be ready by next week?
I'll send out a call announcement if there is sufficient material, otherwise the next call will probably be in early April.
-- Andreas Savva
-- jsdl-wg mailing list jsdl-wg@ogf.org http://www.ogf.org/mailman/listinfo/jsdl-wg
-- Dipl.-Inform. Alexander Papaspyrou http://ds.e-technik.uni-dortmund.de/~alexp Robotics Research Institute phone : +49(231)755-5058 Information Technology Section fax : +49(231)755-3251 Dortmund University of Technology, Germany

Dear All, since our use case is already existing (resource discovery), I guess we could write it up until next week. But since I have an appointment scheduled next week, I would vote for another week. Best regards, Philipp.
Andreas, all,
for the Grid Scheduling use case, I won't make it until next week. As such, I would generally participate *if* there is a call, but not protest if we decide to cancel it. Philipp, what is your opinion on this?
-Alexander
2008/3/13, Andreas Savva <andreas.savva@jp.fujitsu.com>:
Our call schedule is roughly bi-weekly, so the next call would be in two weeks time (Mar.26). But it looks like the chairs won't be available that week.
We are therefore thinking of a call next week, conditional on having sufficient material for discussion. We discussed parameter sweep yesterday and we probably won't have enough material for next week on this topic. That leaves the Activity schema, for which the main action from OGF22 was to provide use cases (Chris, Shabaz, Philipp). So could you please let me know if the use cases are likely to be ready by next week?
I'll send out a call announcement if there is sufficient material, otherwise the next call will probably be in early April.
-- Andreas Savva
-- jsdl-wg mailing list jsdl-wg@ogf.org http://www.ogf.org/mailman/listinfo/jsdl-wg
-- Dipl.-Inform. Alexander Papaspyrou http://ds.e-technik.uni-dortmund.de/~alexp
Robotics Research Institute phone : +49(231)755-5058 Information Technology Section fax : +49(231)755-3251 Dortmund University of Technology, Germany -- jsdl-wg mailing list jsdl-wg@ogf.org http://www.ogf.org/mailman/listinfo/jsdl-wg

Hi, Has there been any thought or discussion on how to represent power usage of a job within the jsdl requirement. Maybe one way to derive the power requirement is from the cpu, memory, network etc. requirments and map it to power models of the execution environment. But has there been any thought of a job explicitly specifying power requirements. Thanks & Regards, Gargi Gargi B Dasgupta IBM India Research Lab Plot No. 4, Phase II, Block C ISID Institutional Area Vasant Kunj, New Delhi 110070 Email:gdasgupt@in.ibm.com Phone:+91 11 51292192

Gargi B Dasgupta wrote:
Has there been any thought or discussion on how to represent power usage of a job within the jsdl requirement. Maybe one way to derive the power requirement is from the cpu, memory, network etc. requirments and map it to power models of the execution environment. But has there been any thought of a job explicitly specifying power requirements.
You mean power as in "quantity measured in Watts"? Assuming that's the case, I'd tend to view it as a quality-of-service (i.e. non-functional) requirement, and also one that is out of our current scope. As such, you'd probably want to use a custom extension term to describe the power requirements, and something like WS-Agreement to drive the enforcement. (OTOH, I don't have access to any sort of system at the moment which could support such a term anyway; all our production kit doesn't know anything about its power consumption.) Hope this helps (a little). Donal.
participants (5)
-
Alexander Papaspyrou
-
Andreas Savva
-
Donal K. Fellows
-
Gargi B Dasgupta
-
Philipp Wieder