I'd like to ask for a feature related to condor_submit.
Every now and then, I have to run a large job cluster with several
100s
to 1000s of jobs, each using $(Process) as the argument to the
executable
(simply a counter, used as a seed for a RNG).
Often I'd like to split the large cluster into pieces submitted
separately,
or extend an existing cluster.
What condor_submit is lacking is a way to start counting at a
process number
0.
Of course, this could be worked around using a "dummy" executable
for the
first n jobs of a cluster:
Executable = /bin/false
Queue n
then defining the real one
Executable = $the_real_one
Queue m
but it imposes unneccessary load on the scheduler, and those very
short running
jobs in the past had the tendency to confuse their shadow processes.
If there was a "StartAt" parameter for the submit file (which would
default
to 0 for the first Queue statement, and be counted up subsequently),
that
would make things a lot easier (also to repeat a selected set of
failed jobs).
Any opinions?
Cheers,
Steffen
--
Steffen Grunewald * MPI Grav.Phys.(AEI) * Am Mühlenberg 1, D-14476
Potsdam
Cluster Admin * http://pandora.aei.mpg.de/merlin/ * http://www.aei.mpg.de/
* e-mail: steffen.grunewald(*)aei.mpg.de * +49-331-567-{fon:7233,fax:
7298}
No Word/PPT mails - http://www.gnu.org/philosophy/no-word-attachments.html
_______________________________________________
Condor-users mailing list
To unsubscribe, send a message to condor-users-request@xxxxxxxxxxx
with a
subject: Unsubscribe
You can also unsubscribe by visiting
https://lists.cs.wisc.edu/mailman/listinfo/condor-users
The archives can be found at:
https://lists.cs.wisc.edu/archive/condor-users/