[kepler-users] Running Kepler workflows on a Torque (OpenPBS) job scheduler

[Jianwu Wang]

Hi Daniel,

    Currently, we already have some support for job scheduler. Norbert 
Podhorszki implemented a set of actors which can submit jobs and check 
job status for the following job scheduler systems: 
PBS/Condor/Fork/LoadLeveler/NCCS/SGE. You can find a sample workflow at 
https://code.kepler-project.org/code/kepler/trunk/workflows/SC06-Tutorial/JobSubmission.xml.  
If you have any questions about it, just let me know.

    Besides the above general mechanism, we also have some actors to 
support some specific job scheduler in cluster or grid systems, such as 
GLite actors, SunGridEngine actors, Nimrod actors, Globus actors.

    We are trying to integrate the above efforts to get a general, 
comprehensive and adaptive mechanism. The corresponding documents will 
also be provided.

Best wishes

Sincerely yours

Jianwu Wang
jianwu at sdsc.edu

Post-Doctor
Scientific Workflow Automation Technologies (SWAT) Laboratory
San Diego Supercomputer Center 
University of California, San Diego
San Diego, U.S.A. 



Chad Berkley wrote:
> Hi Daniel,
>
> We don't currently have any howtos on this.  We have a very 
> prototypical implementation of kepler that will run in distributed 
> environments that has been tested on a ROCKS cluster.  We're in the 
> process of reworking our repository and build system, so I'm not even 
> sure if it works currently.  Jianwu at SDSC has most recently been 
> working on this system so I'll let him jump in on this conversation 
> since he'll have the most up-to-date information.
>
> Here are the links we have for past work, but keep in mind that I'm 
> not sure if any of this will currently work and the build instructions 
> are currently out of date:
>
> https://dev.kepler-project.org/developers/interest-groups/distributed/archive/distributing-nceas-cluster 
>
> https://dev.kepler-project.org/developers/interest-groups/distributed/archive/working-distributed-features 
>
>
> When more information is available on current distributed features, 
> you'll find it here:
> https://dev.kepler-project.org/developers/interest-groups/distributed
>
> We're hoping to get everything stabilized in the next month or so. 
> We've made a large architectural change (modularization) to the core 
> of Kepler so it will take some time.
>
> Thanks,
>
> chad
>
>
> Daniel Korytina wrote:
>> Hi,
>>
>> Can someone please point me to literature/websites that explain how 
>> to run Kepler workflows on standard linux cluster queuing systems, 
>> namely TORQUE or OpenPBS?
>>
>> http://en.wikipedia.org/wiki/TORQUE_Resource_Manager
>> http://en.wikipedia.org/wiki/OpenPBS
>>
>> Thank you,
>> Daniel Korytina
>> _______________________________________________
>> Kepler-users mailing list
>> Kepler-users at kepler-project.org
>> http://mercury.nceas.ucsb.edu/kepler/mailman/listinfo/kepler-users
> _______________________________________________
> Kepler-users mailing list
> Kepler-users at kepler-project.org
> http://mercury.nceas.ucsb.edu/kepler/mailman/listinfo/kepler-users
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.nceas.ucsb.edu/kepler/pipermail/kepler-users/attachments/20090102/93144d58/attachment.html>