Cut3d for large systems

MULTIBINIT, aTDEP, ANADDB, optics, cut3d, mrg* postprocessors

Moderators: MMNSchmitt, gonze

Locked
ouojeda
Posts: 2
Joined: Wed Mar 24, 2010 10:44 pm

Cut3d for large systems

Post by ouojeda » Wed Mar 24, 2010 10:48 pm

Greetings
I've been trying to use cut3d for a system with 48 atoms. It proceeds without issues until reading out the last atom, and it crashes with the following message:
"forrtl: severe (67): input statement requires too much data, unit 19, file /Users/ouo2224/abinit/tatb/eos/e.1/tatbo_VHA"
Is there any way around this?
Thank You
Oscar

mverstra
Posts: 655
Joined: Wed Aug 19, 2009 12:01 pm

Re: Cut3d for large systems

Post by mverstra » Thu Mar 25, 2010 9:12 am

Need a bit more information about what you are actually doing and asking cut3d to do. The VHA file may be corrupted, incomplete, who knows. Give us some platform details, abinit version and so on, and the cut3d commands you are using.

Matthieu
Matthieu Verstraete
University of Liege, Belgium

ouojeda
Posts: 2
Joined: Wed Mar 24, 2010 10:44 pm

Re: Cut3d for large systems

Post by ouojeda » Fri Apr 02, 2010 6:51 pm

Hi
Thank you for your kind reply. I have tired this with the density file also, same result. I've used the intel processor Mac G5 system;
_____________________________________________________________________________________________________
=== Build Information ===
Version : 5.8.3
Build target : i386_darwin9.7.0_intel10.1
Build date : 20090710

=== Compiler Suite ===
C compiler : gnu
CFLAGS : -g -O2
C++ compiler : gnu4.0
CXXFLAGS : -g -O2
Fortran compiler : intel10.1
FCFLAGS : -g -extend_source -vec-report0
FC_LDFLAGS : -i-static -static-libcxa

=== Optimizations ===
Debug level : symbols
Optimization level : standard
Architecture : _

=== MPI ===
Parallel build : yes
Parallel I/O : no
MPI CPPFLAGS : -DMPI=1 -DMPI1=1

=== Linear algebra ===
Library type : abinit
Use ScaLAPACK : no
_____________________________________________________________________

I have also tried the version in an Intel(suse) system, with exactly same output message.
The system has 96 electrons (48 Atoms) and P1 symmetry.
I'm using the "read formatted data option" (1).
When I try to read a single band/kpoint from a wavefunction. I can generate the coordinate system.
Oscar

david.waroquiers
Posts: 138
Joined: Sat Aug 15, 2009 12:45 am

Re: Cut3d for large systems

Post by david.waroquiers » Sat Apr 03, 2010 10:40 am

Hello,

The potential, wavefunction an density files are saved in a binary format. You have to use the option read unformatted binary header + 3d data and not formatted data.

David

jiabo
Posts: 11
Joined: Sat Nov 22, 2014 2:32 am

Re: Cut3d for large systems

Post by jiabo » Mon Dec 08, 2014 12:00 am

Hi Ouojeda,

Have you solved this problem? I think I meet the same problem with you now, can not transfer POT file, if you solved, can you tell me how you make this?

Jiabo

Locked