Different etotal when using multiple nodes

Total energy, geometry optimization, DFT+U, spin....

Moderator: bguster

Locked
mdsimulationgroup
Posts: 13
Joined: Mon Feb 25, 2019 2:20 pm

Different etotal when using multiple nodes

Post by mdsimulationgroup » Wed Feb 27, 2019 11:16 am

Hello,

On a single node, my etotal was -4.4223597790E+04 but when I tried on 2 compute nodes my etotal was -4.4217819731E+04...a 5.78 Ha difference. Any ideas why that would be happening?

Thanks!

ebousquet
Posts: 469
Joined: Tue Apr 19, 2011 11:13 am
Location: University of Liege, Belgium

Re: Different etotal when using multiple nodes

Post by ebousquet » Sat Mar 02, 2019 11:41 am

Dear Stacey,
Is your SCF converged?
Otherwise, did you check the test of your compiled Abinit for parallel (in tests folder, see runtests.py and runparal.py)?
Best wishes,
Eric

Cryst
Posts: 1
Joined: Tue Apr 09, 2019 8:08 pm

Re: Different etotal when using multiple nodes

Post by Cryst » Tue Apr 09, 2019 8:39 pm

Could you shared with link to materials How to check results od Abinit calculations, please?

Thank you

Locked