Hi all,
Currently I'm trying to read the log data (job, user, supervisor and host) for retrieve the resources and few other attributes so that we can track the usage of our server's resource. However I'm stuck in understanding what each column in the log files represent. I've taken a look at administration manual but to no avail.
Is there any document that can guide me into understanding what the value and description in log files(.map, .hst, and few others)?
Just for confirmation, is the log file the best way to get the data that we needed?
Example of the data we need to retrieve are
1) number of cpus used
2) memory used
3) worker node involved
4) user that sumbit the job
5) time spent to execute the job
6) user group
7) job id
8 ) queue time
Another thing, we also have some thinking to use the API, but just wonder if it will become intrusive to the machine's resource?
I'm also not able to find Job Account File(*.acc) in log folder. Is there any setting needed before this file created?
Hopefully can get some advices from you, thanks in advance.
