From: Axel Kohlmeyer (akohlmey_at_gmail.com)
Date: Thu Nov 12 2009 - 07:36:00 CST
2009/11/12 Ming <ebullience_at_emails.bjut.edu.cn>:
> Dear all,
>
> I am using the NAMD_2.6_Linux-ia64-MPT version on a SGI Altix 64 cpu
> workstation. I've set the MPI_DSM_DISTRIBUTE=yes and MPI_MEMMAP_OFF=1 in my
> .bashrc. The job can ban successfully now but I found the memory keep
> increasing when the job runing. Below is part of the output from the "free
> -m -s 10" commond. Is this normal? Any suggestions? Thanks in advance!
whether you have a memory leak or not is very hard to tell.
please note, that if you are equilibrating a system, particularly
and input fresh out of "solvate", then your neighbor lists may
be growing for a while. you have to monitor for a long time, since
most codes only increase their internal memory pools and never
shrink them. at some point the memory use should level out.
you also have to check what else is running on the machine.
in any case, you still have plenty of memory left. so why worry?
cheers,
axel.
>
>
> Mem: 61521 6114 55407 0 0 461
> -/+ buffers/cache: 5652 55869
> Swap: 9999 0 9999
>
> total used free shared buffers cached
> Mem: 61521 6151 55369 0 0 462
> -/+ buffers/cache: 5689 55831
> Swap: 9999 0 9999
>
> total used free shared buffers cached
> Mem: 61521 6184 55337 0 0 462
> -/+ buffers/cache: 5721 55799
> Swap: 9999 0 9999
>
> total used free shared buffers cached
> Mem: 61521 6188 55333 0 0 462
> -/+ buffers/cache: 5725 55795
> Swap: 9999 0 9999
>
> total used free shared buffers cached
> Mem: 61521 6191 55330 0 0 463
> -/+ buffers/cache: 5728 55793
> Swap: 9999 0 9999
>
> total used free shared buffers cached
> Mem: 61521 6200 55321 0 0 463
> -/+ buffers/cache: 5736 55784
> Swap: 9999 0 9999
>
>
-- Dr. Axel Kohlmeyer akohlmey_at_gmail.com Institute for Computational Molecular Science College of Science and Technology Temple University, Philadelphia PA, USA.
This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 15:53:29 CST