Re: [Apbs-users] Crash

[Nathan B. <ba...@bi...>]

This sounds like a memory limitation on your machine; the memory usage 
of PMG, the underlying multigrid solver, scales as ~160 B/grid point. 
I'd try using asynchronous parallel focusing for this to break the 
calculation into a series of smaller runs which can be performed 
sequentially.  There is information on this in the manual.

Thanks,

Nathan

Frank D Ducheneaux wrote:
> Hi Dr. Baker,
> 
> For reasons I will omit, I am trying to do some electrostatic calculations
> on a series of proteins (~5000-6000 atoms).  I first use your psize tool
> to get the x,y,z lengths.  I then take the longest axis and do several
> calculations on one molecule with the square box dimensions all a certain
> percentage of this longest axis.  One protein in question had its longest
> length at ~ 70 A and I tried to a calculation with the square box
> dimensions 3X this length, or 210 A.  I wanted to have a significant
> amount of grid points but the calculation failed until I brought the dime
> down to 129 129 129.  The failure was :
> 
> CALCULATION #1 (solvated_state): MULTIGRID
>   Setting up problem...
>   Vpbe_ctor:  Using max ion radius (2 A) for exclusion function
> VASSERT: ASSERTION FAILURE!  filename vmem.c, line 245, (ram != ((void
> *)0))
> Abort (core dumped)
> 0.910u 0.440s 0:05.70 23.6%     0+0k 0+0io 298pf+0w
> 
> 
> This was a linear calculation as there were no counter ions.  I am
> assuming that this might have been a memory failure, but i wanted to make
> sure.  Any insights as to how I might get around this and successfully
> complete this over-the-top calculation?  Also, it appears that the
> calculation is not letting us set nlev to any value but 4.  Thanks in
> advance.
> 
> 
> "People always tell you to 'rise above it' and 'don't look down on
> other people'.  How are you supposed to do both?"
> 
> 
> Frank Ducheneaux
> California Institute of Technology
> Division of Chemistry
> M/C 127-72
> Pasadena, CA 91125
> 
> 626-395-2689 (office)
> 626-226-6960 (home/cell)
> 
> 
> _______________________________________________
> apbs-users mailing list
> apb...@ch...
> http://cholla.wustl.edu/mailman/listinfo/apbs-users

-- 
Nathan A. Baker, Assistant Professor
Washington University in St. Louis School of Medicine
Dept. of Biochemistry and Molecular Biophysics
Center for Computational Biology
700 S. Euclid Ave., Campus Box 8036, St. Louis, MO 63110
Phone:  (314) 362-2040, Fax:  (314) 362-0234
URL:  http://www.biochem.wustl.edu/~baker