[Wien] Problems with hybrid calculation

[Pavel Ondracka]

On Tue, 2015-03-03 at 17:09 +0900, Paul Fons wrote:
> Hi I did run init_hf_lapw and I saw no sign of errors upon running it,
> however, I ran it from the web interface the first time.  I just
> restarted the calculation using bash and have confirmed the ibz and
> fbz files exist.
> The reason that I am interested in hybrid mode is that I actually have
> 96 processors available with infiniband interconnects.  Is a 96 atom
> system too expensive to run with PBE0 even with 96 cores? How much
> time should I expect such a calculation to take?
> 
> 
> Paul Fons 

I've never run such a big hybrid calculation, however I've stumbled upon
article about hybrid calculations for such a system (96 atoms am-HfO2
[1]) and they used a BlueGene cluster with 18TFLOPS output (according to
TOP500 site). 
Its not clear from the article how big part of the system they actually
used, but this should give you some ideas about the cost. 

[1] Broqvist, P. & Pasquarello, A. Band gaps and dielectric constants of
amorphous hafnium silicates: A first-principles investigation. Appl.
Phys. Lett. 90, 082907 (2007).