Page 1 of 1

VASP and EON

Posted: Wed Dec 02, 2015 2:15 pm
by rafaelbna
Dear,

My name is Rafael, I am a PhD at the Uppsala University and a new user of EON.
I would like to perform some basin hopping calculation with vasp as the potential. However, I am facing some problem.

I would really appreciate if you could kindly give me some advices to overcome these issues.

I am trying to perform basin hopping calculation using VASP as the potential for a small Hydrogen cluster to make it simple. My problems are:

1. Using INCAR coming for example akmc-au-vasp vasp calculation never converge. If I comment EDIFFG tag then it converges but I get the error message 'vasp died unexpectedly!'. I have attached here the files I used for this test.

2. I am also trying the example akmc-au-vasp. I am getting the following error:
'IndentationError: expected an indented block'



I am looking forward to hearing from you,

kind regards,
Rafael

Re: VASP and EON

Posted: Mon Apr 03, 2017 8:53 pm
by mirhosseini
I also facing the same problem. I can run the akmc-au-vasp example with some potential but not with vasp and mpi.
Could you please comment on this issue?
Thanks

Re: VASP and EON

Posted: Tue Apr 04, 2017 3:28 pm
by graeme
If you can post a .tar.gz file of the calculation so that I can reproduce the problem, then I'll also try to debug it.

I see that the example in the akmc-au-vasp directory uses the mpi potential. Using this is a bit complicated, and it requires rebuilding vasp with our mpmd modification, included with the vtstcode. I suggest using the regular vasp potential and the cluster communicator. In fact, we should really have that setup in our example. I will add one this week. I will also check about the indentation error.

Re: VASP and EON

Posted: Wed May 03, 2017 6:18 pm
by rdgoliveira
Is there any release date for this example with Vasp potential and cluster communicator?

Thank you

Re: VASP and EON

Posted: Sun May 07, 2017 5:34 pm
by graeme
The attached example runs akmc through eon with the vasp calculator and the cluster communicator. You will have to change the system specific commands including how to submit and delete jobs, query the queue, and run vasp. I was planning to make a faster example, but this is one calculation that I had on hand. It is also included in the eon code.

Re: VASP and EON

Posted: Wed Jul 05, 2017 7:31 am
by mirhosseini
Dear Graeme,
Thanks for the example.
I am not still able to run this example. When I call eonclient I got this error message:

Loading parameter file config.ini
* [Main] job: akmc
* [Main] temperature: 300
* [Main] random_seed: 500
* [Main] checkpoint: true
* [Main] finite_difference: 0.1
* [Potential] potential: vasp
* [Debug] write_movies: True
* [Structure Comparison] distance_difference: 0.2
* [Structure Comparison] neighbor_cutoff: 3.4
* [Structure Comparison] energy_difference: 0.1
* [Structure Comparison] indistinguishable_atoms: true
* [Optimizer] opt_method: lbfgs
* [Optimizer] converged_force: 0.02
* [Optimizer] max_iterations: 250
* [Prefactor] default_value: 1e12
* [Parallel Replica] dephase_time: 250
* [Saddle Search] method: dynamics
* [Saddle Search] dynamics_temperature: 800
error: Unknown job: akmc

I used the same config file and I changed the job script according to our system .
Could you please comment on that?

Kind regards,
Hossein

Re: VASP and EON

Posted: Wed Jul 05, 2017 1:19 pm
by graeme
Make sure that you run "eon" and not "eonclient" when running an akmc job.