[Wien] Space Group 204:92 atoms per unit cell

[Ghosh SUDDHASATTWA]

Dear Wien2k users, 

I have generated a Cd intermetallic compound with Space Group 204 with 92
atoms per unit cell. We have also configured our cluster with k-point
parallel technique. Is their any way out to know a priori as to the minimum
number of cpu's required to finish the SCF with in1new switch within a
reasonable time period. 

I did a recent experiment on a primitive monoclinic lattice and would like
to know if anything can be concluded anything out of this. 

I used the k-point parallel technique. 

 

I had a 16 atom unit cell with a Space Group 11. 

I started the calculation with 16 cpu's and 32 cpu's 

The calculation with 16 cpu's I thought finished faster than the latter one.
Does it mean that a 92 atom cell would require 92 cpu's. How to find out the
minimum number of cpu's to finish the SCF in a reasonable time? 

 

Any suggestions 

 

Suddhasattwa 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://zeus.theochem.tuwien.ac.at/pipermail/wien/attachments/20100702/bbb862a5/attachment.htm>