2 errors when doing HSE calculation

Problems running VASP: crashes, internal errors, "wrong" results.


Moderators: Global Moderator, Moderator

Post Reply
Message
Author
alirezaa66
Newbie
Newbie
Posts: 5
Joined: Mon Sep 17, 2012 11:21 pm
License Nr.: 5-254

2 errors when doing HSE calculation

#1 Post by alirezaa66 » Tue Apr 02, 2013 2:30 am

I am doing Hartree Fock part of the HSE calculation and I got this error at first:
'LAPACK: Routine ZPOTRF failed!',INFO,NK,ISP

which is coming from Choleski2.F
After doing all the suggestion made in this forum about this error I am still getting either the same error or this:

'EDWAV: internal error, the gradient is not orthogonal',NK,NP,DORT

which is coming from rot.F

I understand that orthogonality of anything might be related to my input; however, the problem is I run the same calculation on another cluster and it successfully finishes but when I try it again on Stampde (one of TACC clusters) I get the same error! So I guess my inputs are fine. I have tried installing LAPACK and using it instead of Stampede LAPACK too and it didn't work. Any suggestion will be appreciated, please help!!
Last edited by alirezaa66 on Tue Apr 02, 2013 2:30 am, edited 1 time in total.

admin
Administrator
Administrator
Posts: 2921
Joined: Tue Aug 03, 2004 8:18 am
License Nr.: 458

2 errors when doing HSE calculation

#2 Post by admin » Mon Apr 08, 2013 6:16 pm

which vasp-release do you use?
maybe you have started from an incompatible WAVECAR?
Last edited by admin on Mon Apr 08, 2013 6:16 pm, edited 1 time in total.

alirezaa66
Newbie
Newbie
Posts: 5
Joined: Mon Sep 17, 2012 11:21 pm
License Nr.: 5-254

2 errors when doing HSE calculation

#3 Post by alirezaa66 » Tue Apr 16, 2013 8:22 pm

Thanks admin for your reply; I tried 5.3.3 and 5.2.12 but the issue has been resolved with help of TACC. They tested my calculation with a vasp 5.3.3 compiled with intel_12 compiler and it worked! so it might be useful information for people who want to do HSE, intel_13 compiler might cause some numerical instability in the steps that Hartree-Fock calculations are done. So this was NOT VASP issue.
Last edited by alirezaa66 on Tue Apr 16, 2013 8:22 pm, edited 1 time in total.

Post Reply