<< SIESTA example working >>

[Camps]

Hello all,

Does any one could send me a SIESTA input file that run properly?

I had been working for two days setting up the T6_SIESTA_Carbon
example unsuccessfully.

The last info in the log file was:

>>
notYet =

1

Best regards,

Camps

PS: I read the USPEX manual and all the Ti (i=0..6) examples, too.

[Andriy Lyakhov]

Hello,

the message you get hints that file still_reading was not deleted
properly from the previous run. Just delete it manually before
starting the USPEX again.

Also, to answer one of your previous questions - please make sure that
files getStuff and getFromTo are:

(1) present in the main folder, in the FunctionFolder and Specific.
(2) labeled as executable.

These files seem to be missing from the T6_SIESTA main folder. Please
copy those files and create the folder Seeds if you want to run the T6
example. Copying to the folder Code will also work, if you don't
forget to copy Specific (exactly what you did).

Now, about SIESTA run itself, there are few things you have to keep in
mind:

1) Make sure that your SIESTA works properly and the second line in
commandExecutable parameter in INPUT_EA.txt is actually launching the
SIESTA code for you. If you use compiled single file and launch it
with ./siesta, then just go into CalcFold1 and manually type the
following line:
./siesta < input.fdf>output

If it runs flawlessly and produces output - fine. If not, you will
have to fix SIESTA before running USPEX again. For example, if you
just use siesta from the T6_SIESTA example (folder Specific), then it
is very probable that you won't have some libraries to run it. Thus
I'll advise you to compile siesta by yourself or make sure that
folders with all the required fortran libraries are specified in
your .bashrc file.

(2) Assuming that siesta code runs fine for you, USPEX should also
work fine. However unlike other test cases, it works in parallel
regime SIESTAlocal instead of nonParallel one. Therefore you can see 6
calculation folders instead of 1. The parallel regime is similar to
working on supercomputer and thus it is important to understand how it
works. USPEX will launch, submit 6 jobs and then quit. When you launch
it again, it checks whether jobs were finished, and in this case it
starts new jobs and quits. If they were not finished yet, USPEX just
quits without doing anything. Therefore it's important to launch the
USPEX regularly using the crontab routine. For T6 example you can
safely run the USPEX every few minutes, while on supercomputers the
intervals can be as big as several hours.

If you are familiar with crontab - use the command like:
0-59/4 * * * * cd the path of your directory/ && ./job

where job is executable that contains the line:
matlab <ev_alg.m >doedl

If you are not familiar with crontab and just want to try how USPEX
works with siesta, then you may change the following parameters in the
input file:

1 : numParallelCalcs (how many parallel calculations shall be
performed)
nonParallel : whichCluster
remove the & at the end of the second line in commandExecutable (this
is important!)

This will launch USPEX in 'familiar' non parallel mode, where only 1
job at a time is being executed and you don't have to restart it
regularly (assuming that everything goes well).

If you have any other questions or comments - don't hesitate to ask.

Sincerely,
Andriy

[Muthukumar Kaliappan]

Hello all,

I get the following error message at generation 2.

If i restart the calculation from generation 2, the 3 generation runs fine,

but i still get the same error message at the end of the 3 generation.

Could you please let me know where it is going wrong.

My best regards,

Muthu

Error in ==> move_all_atom_Mutation at 2

global ORG_STRUC

??? Output argument "new_Coord" (and maybe others) not assigned during call to

"/scratch/agvalenti/muthukumar/USPEX_New/QZ_Calc_Parameter/WC_Seed_Technique/Check_New/FunctionFolder/move_all_atom_Mutation.m>move_all_atom_Mutation".

Error in ==> atomMutation at 16

[MUT_COORD]=

move_all_atom_Mutation(POP_STRUC.POPULATION(ind).COORDINATES,POP_STRUC.POPULATION(ind).numIons,POP_STRUC.POPULATION(ind).LATTICE,

POP_STRUC.POPULATION(ind).order);

 

^G

log2 (END) 

--
Muthukumar Kaliappan

Room 1.144
Institute for Theoretical Physics
J. W. Goethe University
Max von Laue str. 1

60438 Frankfurt am Main

Germany

--------------

எந்த செயலையும் முடியவில்லை அல்லது சரியாக செய்ய முடியாது, அல்லது வெற்றி காண
முடியாது என்று எண்ணி துவங்காதீர்கள். நம்பிக்கையை வளர்த்துக் கொள்ளுங்கள்.
எடுத்த செயலில் வெற்றிக்காக போராடுங்கள். லட்சியத்துடன் அதனை அடைவதற்கான
முயற்சியை அதிகப்படுத்துங்கள்.

---------------