[Wien] Can I reduce size of vector files in band calculation

Peter Blaha peter.blaha at tuwien.ac.at
Thu Jul 11 20:18:48 CEST 2024


2 possibilities:
If you need only valence bands, you can reduce  EMAX in case.inso to a 
value a bit above EF.
If your case has many semicore states, you can increase EMIN in case.in1 
(you will get "XX eigenvalues below ..") to cut these states off. Set it 
carefully into a "gap" region.

PS: A decrease of EMAX in case.in1 will reduce the accuracy of the SO 
eigenvalues.

PPS: Depending on what you really need (just eigenvalues ?) you could 
also split your k-mesh into some parts and concatenate the energyso or 
outputso files  (like   x lapw1 -up/dn -p; x lapwso -up -p with some 
.machines files (with hostnames which do not exists). The job will 
crash, but you can then execute
lapw1(c) uplapw1_1.def
lapw1(c) dnlapw1_1.def
lapwso uplapwso_1.def

remove the vector files and do the same with _2, _3 ...

x spaghetti -so -up -p will take the the outputso_* files and 
concatenate them.

Am 11.07.2024 um 17:56 schrieb pluto via Wien:
> Dear All,
> 
> I am calculating bands for a mesh of k-points for a big slab (e.g. 51x51 
> kpoints)
> 
> Vector files become very large during such band structure calculation 
> and I am having disk space issues.
> 
> Is there a way to reduce the size of vector files? It is FM+SOC 
> calculation, so I cannot separately calculate up/dn spins.
> 
> I only need valence bands, say down to 10 eV binding.
> 
> Best,
> Lukasz
> 
> _______________________________________________
> Wien mailing list
> Wien at zeus.theochem.tuwien.ac.at
> http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
> SEARCH the MAILING-LIST at:  
> http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html

-- 
-----------------------------------------------------------------------
Peter Blaha,  Inst. f. Materials Chemistry, TU Vienna, A-1060 Vienna
Phone: +43-158801165300
Email: peter.blaha at tuwien.ac.at
WWW:   http://www.imc.tuwien.ac.at      WIEN2k: http://www.wien2k.at
-------------------------------------------------------------------------


More information about the Wien mailing list