[Wannier] pw2wannier very slow on large system

[Mostofi, Arash]

Hi Daniel

Whilst k-point parallelisation is not implemented in pw2wannier90, FFT grid parallelisation is. The latter should provide a speedup for a large supercell (which I’d normally expect to have few k-points but a large FFT grid).

As an aside, we do ask that everyone signs with their full name and affiliation on posts to this forum.

Best wishes,

Arash

—
Professor Arash Mostofi — www.mostofigroup.org<http://www.mostofigroup.org>
Director, Thomas Young Centre @Imperial
Imperial College London

On 16 Nov 2020, at 19:07, Daniel Marchand <t-dmarchand at microsoft.com<mailto:t-dmarchand at microsoft.com>> wrote:

Hi Everyone,

We’re having some trouble running pw2wannier on a very large system as it is running very slow. To give some comparison, we are able to finish the nscf calculation in ~4hours, while pw2wannier has not finished the first MMN: iknum step even after ~8 hours of runtime. It seems like in theory pw2wannier should be parallelizable but we found poor performance beyond a single node.

Is there anything we can do to help speed up the process? Parallelization steps to try out or common pitfalls that we could avoid?

Best,

Daniel
_______________________________________________
Wannier mailing list
Wannier at lists.quantum-espresso.org<mailto:Wannier at lists.quantum-espresso.org>
https://lists.quantum-espresso.org/mailman/listinfo/wannier

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/wannier/attachments/20201118/09b55747/attachment.html>