[Wien] dstart.error

[Peter Blaha]

As was mentioned, such simulations are possible on a bigger cluster with 
infiniband connection and at least 64 (better more) cores in total.

However, unless you already have sufficient experience with WIEN2k, I 
would STRONGLY suggest to "learn" how to run such big calculations on 
eg. a 2x2 supercell first.
I'm pretty sure, you can reproduce their results (and experiment) also 
with a smaller cell. Next step would be to do a 3x3 cell and compare to 
your previous results. If nothing changes, you are ok.

Regards
On 9/8/20 10:46 PM, Ilias Miroslav, doc. RNDr., PhD. wrote:
> Ad: 
> https://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/msg20439.html
> 
> Dear Professor Blaha,
> 
> thanks for your answer.
> 
> I took this Quartz structure from the work of Pershina, 
> https://pubs.rsc.org/en/content/articlepdf/2016/cp/c6cp02253g , where 
> the 4x4 supercell is the best choice for comparison with experiments.
> 
> What could be done to make such 4x4 supercell  feasible with Wien2k ? 
> This is slightly more than 300 atoms, no symmetry (H-passivation is only 
> from the top side).  Clearly, I would reduce the vacuum to 40 Bohr, as 
> you suggest, but  would it be sufficient ? If there is a hope to 
> simulate such system with Wien2k on 256GB RAM equipped nodes, I would 
> try it...
> 
> Best, Miro
> 
> 
> _______________________________________________
> 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
> 

-- 

                                       P.Blaha
--------------------------------------------------------------------------
Peter BLAHA, Inst.f. Materials Chemistry, TU Vienna, A-1060 Vienna
Phone: +43-1-58801-165300             FAX: +43-1-58801-165982
Email: blaha at theochem.tuwien.ac.at    WIEN2k: http://www.wien2k.at
WWW:   http://www.imc.tuwien.ac.at/TC_Blaha
--------------------------------------------------------------------------