AMBER Archive (2009)
Subject: Re: [AMBER] addles segmentation fault
From: xiaonan zhang (heptoking_at_gmail.com)
Date: Mon Mar 09 2009 - 08:08:47 CDT
Prof. Simmerling I was sort of worried about the suitability of this
big system for NEB. Especially, at present the whole
system should be replicated. But I actually tried 6 copies but the result
was the same.
So maybe my system is really too big for this.
Thank you for your advice.
On Mon, Mar 9, 2009 at 7:45 PM, Carlos Simmerling <
> it's possible that you are creating an array that is too large
> somewhere, such as the exclusion list which will be huge in this case.
> it may not be very efficient- for example, each of your 8596*30 atoms
> will have 8596*29 atoms in the exclusion list using LES NEB,
> 8596*30*8596*29=64 million excluded atoms. you might try making things
> smaller (fewer neb beads) and see if it helps. I'm not sure if anyone
> has tried a system this large using LES NEB.
> > Dear Amber users
> > I am trying to do some nudged elastic banding simulation on a big
> protein (548 residues).
> > I just went through this by reading the NEB tutorial, which is quite
> informative and helpful.
> > (Many thanks to Dr. Ross Walker)
> > However, when I used addles to create 30 copies of my protein. The
> output file said
> > Checking topology sizes against compiled limits
> > exceeded MAXANG in SIZE.BLOCK.
> > MAXANG = 8000
> > ntheth = 9788
> > exceeded MAXDIH in SIZE.BLOCK.
> > MAXDIH = 15000
> > nphih = 18572
> > Change SIZE.BLOCK and recompile.
> > I searched the mailing list and found the SIZE.h file and increased these
> values to 10000 and 20000 respectively to fit my protein.
> > after make clean,
> > and make in the $AMBERHOME/src/addles/ folder
> > I created a new addles program in this folder.
> > But when I used this command in this folder to create the prmtop and
> inpcrd file.
> > There came a segmentation fault
> > forrtl: severe (174): SIGSEGV, segmentation fault occurred
> > Image PC Routine Line Source
> > addles 08051E1A Unknown Unknown Unknown
> > addles 0804CD72 Unknown Unknown Unknown
> > addles 08049B95 Unknown Unknown Unknown
> > libc.so.6 0076EE33 Unknown Unknown Unknown
> > addles 08049AD1 Unknown Unknown Unknown
> > The input and output files are attached.
> > I really don't know the reason for this.
> > Please give some suggestions.
> > Best Regards!
> > Xiaonan Zhang
> > Shanghai Public Health Clinical Center
> > _______________________________________________
> > AMBER mailing list
> > AMBER_at_ambermd.org
> > http://lists.ambermd.org/mailman/listinfo/amber
> AMBER mailing list
AMBER mailing list