[gpaw-users] bug in Mixer with spinpol=True ?

[Nichols A. Romero]

Hi,

This is from another user at ANL. It ran on 1 node (8-cores) of a Linux cluster.
The calculation is for spinpol=True.

mixer = Mixer(0.10, 5, 100.0) leads to an out-of-memory error

mixer = MixerSum(0.10, 5, 100.0) works

This is surprising since a --dry-run memory estimate does not show any
memory usage difference between using Mixer or MixerSum.

Does anyone know what might be going on?

-- 
Nichols A. Romero, Ph.D.
Argonne Leadership Computing Facility
Argonne National Laboratory
Building 240 Room 2-127
9700 South Cass Avenue
Argonne, IL 60490
(630) 252-3441

-------------- next part --------------
A non-text attachment was scrubbed...
Name: ag1_plat6.py
Type: text/x-python
Size: 534 bytes
Desc: not available
Url : http://listserv.fysik.dtu.dk/pipermail/gpaw-users/attachments/20110125/6fd33795/attachment.py 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: POSCAR
Type: application/octet-stream
Size: 471 bytes
Desc: not available
Url : http://listserv.fysik.dtu.dk/pipermail/gpaw-users/attachments/20110125/6fd33795/attachment.obj