Hi Ryan,
Not sure what happened. As the threads are handled directly through
pthreads, I would be very surprised if a job with --j 1 actually launched
5 threads...
Sjors
> Never did get a response to this one -- I hope that someone can tell me if
> I just asked a bad question or something.
>
> In any case, some followup: I took a look at my user's job. He has
> requested --j 1, but when I look at /proc/<PID>/status of each of his
> relion_refine_mpi processes (grep -i thread /proc/<PID>/status):
>
> 27472: Threads: 4
> 27473: Threads: 5
> 27474: Threads: 5
> 27475: Threads: 5
> 27476: Threads: 5
> 27477: Threads: 5
> 27478: Threads: 5
> 27479: Threads: 5
> 27480: Threads: 5
> 27481: Threads: 5
> 27482: Threads: 5
> 27483: Threads: 5
> 27484: Threads: 5
> 27485: Threads: 5
> 27486: Threads: 5
> 27487: Threads: 5
> 27488: Threads: 5
> 27489: Threads: 5
> 27490: Threads: 5
> 27491: Threads: 5
> 27492: Threads: 5
> 27493: Threads: 5
> 27494: Threads: 5
> 27495: Threads: 5
>
> Does anyone know why this would be, or how one should appropriately run
> one thread per core? Shouldn't this be hurting performance?
>
--
Sjors Scheres
MRC Laboratory of Molecular Biology
Francis Crick Avenue, Cambridge Biomedical Campus
Cambridge CB2 0QH, U.K.
tel: +44 (0)1223 267061
http://www2.mrc-lmb.cam.ac.uk/groups/scheres
|