[gmx-users] genbox job killed

[David Mobley]

To elaborate on David's response, your real problem is that you have 279936
waters in your system, which is over a million atoms. This is a ridiculusly
large number, so either you are using a box size that is WAY too big, or
your protein is VERY stretched out. For example, I have a 130 amino acid
protein which I solvate with something like 6,000 waters for a total number
of atoms in the neighborhood of 20,000.

If your computer didn't run out of memory, you'd probably be e-mailing the
list asking why your system was running so slow. If you're running
calculations on a system with a million atoms it will be ridiculously slow.
(With 20,000-ish atoms I get something like 36 node-hrs/ns for my system,
and processor time doesn't scale that well with number of atoms).

David Mobley


On 12/22/05, Wang Zhun <wangzhun at pumc.edu.cn> wrote:
>
> Hi,
>
> Genbox command works well for my previous application: solvation a peptide
> of 20 AAs on a Compaq AP400 working station(PII400x2,128M).
> But when I turned to the present job:solvating a protein of 235 AAs, the
> job was killed. It reads like as below:
> Initialising van der waals distances...
> Will generate new solvent configuration of 12x12x9 boxes
> Generating configuration
> Sorting configuration
> Found 1 molecule type:
>     SOL (   3 atoms): 279936 residues
> Calculating Overlap...
> box_margin = 0.315
> Removed 85908 atoms that were outside the box
> Table routines are used for coulomb: FALSE
> Table routines are used for vdw:     FALSE
> Cut-off's:   NS: 0.48   Coulomb: 0.48   LJ: 0.48
> Generated table with 0 data points for COUL.
> Tabscale = 500 points/nm
> Generated table with 0 data points for LJ6.
> Tabscale = 500 points/nm
> Generated table with 0 data points for LJ12.
> Tabscale = 500 points/nm
> Going to determine what solvent types we have.
> There are 0 molecules, 1006805 charge groups and 1006805 atoms
> There are 0 optimized solvent molecules on node 0
> There are 0 optimized water molecules on node 0
> Grid: 90 x 90 x 63 cells
> Killed
>
>
> What's the problem?
>
>
> _______________________________________________
> gmx-users mailing list
> gmx-users at gromacs.org
> http://www.gromacs.org/mailman/listinfo/gmx-users
> Please don't post (un)subscribe requests to the list. Use the
> www interface or send it to gmx-users-request at gromacs.org.
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://maillist.sys.kth.se/pipermail/gromacs.org_gmx-users/attachments/20051223/547aeb44/attachment.html>