Dallas B. Warren wrote:
Is there something that can do to force the files to be written to the disk?
How the operating system manages the buffering is something you'd
usually have to talk to the OS about.
I am trying to work out why a simulation on a SC is not doing what it should.
Problem is, only the very start of the files are present, since they are not
large enough to be written, and the job reaches it's walltime (longer walltime
does not good) and is killed.
Might as well put here what the issue is too, someone might have an idea. I
have contacted the SC people, but they are yet to get back to me.
Have a simulation that runs fine on my own computer that is completed within
1.25 hr. Using all the exactly fines and execution files (except they are
within a script on the SC), put it on SC, run the job, and it is killed with a
10 hr walltime. Something is majorly wrong, just trying to work out how I can
look into what is going on, since only have initial output to look at. Last
thing in the log file is:
Well, it's hanging after starting successfully, but we can't really
guess why without some more information on the hardware, compilation
method and simulation details.
Mark
_______________________________________________
gmx-users mailing list [email protected]
http://www.gromacs.org/mailman/listinfo/gmx-users
Please search the archive at http://www.gromacs.org/search before posting!
Please don't post (un)subscribe requests to the list. Use the
www interface or send it to [email protected].
Can't post? Read http://www.gromacs.org/mailing_lists/users.php
