Dear all,
When doing an excitonic absorption spectrum calculation, Yambo creates a 06_BSE directory. This directory can occupy really a lot of space. I was wondering what it contains exactly, and how important it is to keep it. Could you please comment on this?
Best regards,
Rolando
P.S. Regarding the abinit version, one that poses no problem is 5.4.4p. According to the Abinit web site, it should be the preferred production version. I tried version 5.7.3, but had the same problem as with version 5.7.2.
On absorption spectra
Moderators: Davide Sangalli, andrea.ferretti, myrta gruning, andrea marini, Daniele Varsano, Conor Hogan
-
- Posts: 5
- Joined: Tue Mar 31, 2009 2:53 pm
Re: On absorption spectra
**********************
Departement Fysica
Universiteit Antwerpen
Groenenborgerlaan 171
B-2020 Antwerpen
Belgium
Tel. (32)(3)265.35.45
Departement Fysica
Universiteit Antwerpen
Groenenborgerlaan 171
B-2020 Antwerpen
Belgium
Tel. (32)(3)265.35.45
- myrta gruning
- Posts: 242
- Joined: Tue Mar 17, 2009 11:38 am
- Contact:
Re: On absorption spectra
Dear Rolando
usually what I got from a BSE calculation in the SAVE directory is the database (n)db.BS_Q1, or in case I use the -S option (DataBases fragmentation) I got BS_Q1_*_* directories, each containg a ndb.fragment. Probably you are then using the -J option (Job string identifier) followed by 06_BSE when calling yambo?
Assuming this is the case (so what happen is that yambo use the 06_BSE directory to store data instead of the default SAVE), this directory contains the BS matrix. It may indeed be quite large depending on the valence/conduction bands/k points involved.
To keep it or not depends on you.
Note that you have two stages: the first is the calculation of the BS matrix, the second is the solution of the eigenvalue problem involving this matrix. For the solution usually the Haydock method is used that calculates directly the spectrum. If you would like a spectrum for a different range of energies, or a different broadening then only the second stage has to be repeated, while if you delete the BSE matrix you have to repeat the whole calculation. From the log file you can have an idea on the time it cost to calculate each part.
So keep it or not depends if you want to reuse the data or not, and on how much does it cost to keep the data .vs. how much does it cost to rerun the calculation.
About converting the latest abinit (5.7.*) KSS which revision of yambo are you using?
Cheers
m
usually what I got from a BSE calculation in the SAVE directory is the database (n)db.BS_Q1, or in case I use the -S option (DataBases fragmentation) I got BS_Q1_*_* directories, each containg a ndb.fragment. Probably you are then using the -J option (Job string identifier) followed by 06_BSE when calling yambo?
Assuming this is the case (so what happen is that yambo use the 06_BSE directory to store data instead of the default SAVE), this directory contains the BS matrix. It may indeed be quite large depending on the valence/conduction bands/k points involved.
To keep it or not depends on you.
Note that you have two stages: the first is the calculation of the BS matrix, the second is the solution of the eigenvalue problem involving this matrix. For the solution usually the Haydock method is used that calculates directly the spectrum. If you would like a spectrum for a different range of energies, or a different broadening then only the second stage has to be repeated, while if you delete the BSE matrix you have to repeat the whole calculation. From the log file you can have an idea on the time it cost to calculate each part.
So keep it or not depends if you want to reuse the data or not, and on how much does it cost to keep the data .vs. how much does it cost to rerun the calculation.
About converting the latest abinit (5.7.*) KSS which revision of yambo are you using?
Cheers
m
Dr Myrta Grüning
School of Mathematics and Physics
Queen's University Belfast - Northern Ireland
http://www.researcherid.com/rid/B-1515-2009
School of Mathematics and Physics
Queen's University Belfast - Northern Ireland
http://www.researcherid.com/rid/B-1515-2009
- myrta gruning
- Posts: 242
- Joined: Tue Mar 17, 2009 11:38 am
- Contact:
Re: On absorption spectra
One more thing: About the problems with the KSS post it in the forum <<Importing core databases (a2y, p2y and e2y)>>. I just copied the error you signaled there, in a new topic.
thanks,
m
thanks,
m
Dr Myrta Grüning
School of Mathematics and Physics
Queen's University Belfast - Northern Ireland
http://www.researcherid.com/rid/B-1515-2009
School of Mathematics and Physics
Queen's University Belfast - Northern Ireland
http://www.researcherid.com/rid/B-1515-2009