sPHENIX discussion of software

[pinkenburg <pinkenburg AT bnl.gov>]

Hi folks,

as announced in the software meeting - we need to change our condor 
habits a bit. We are putting more load on our cluster and need to make 
sure that essential jobs have the resources to run while also getting 
analysis done. It's basically a repeat of what we had in our example 
condor job file. For those who read and implemented it already, you just 
need to add the concurrency limit, for others who use handme-down job 
files, please have a look at yours:

https://wiki.sphenix.bnl.gov/index.php?title=Condor#Mandatory_entries_in_your_condor_job_file_until_further_notice

This doesn't mean you have to terminate the jobs currently running but 
for the next submission - please make those changes. We will see how 
this goes - there will likely be adjustments to the maximum number of 
jobs which can run (hopefully we can run more).

When we run into a condor crunch I do look at running jobs. Jobs which 
do not follow the above will be terminated first to make room. I hope it 
won't come to that, I'd rather see everything running smoothly. After 
all we are in this together.

Chris

--
*************************************************************

Christopher H. Pinkenburg       ;    pinkenburg AT bnl.gov
                                ;    http://www.phenix.bnl.gov/~pinkenbu

Brookhaven National Laboratory  ;    phone: (631) 344-5692
Physics Department Bldg 510 C   ;    fax:   (631) 344-3253
Upton, NY 11973-5000

*************************************************************