Replies: 2 comments 1 reply
-
Did you launch the code using mpirun or mpiexec? |
Beta Was this translation helpful? Give feedback.
1 reply
-
I have the following settings for my calculations. But the time it takes is
very long for geometry optimization.
Can you suggest the best setting for the fastest calculation for the system
given below?
…On Mon, Aug 5, 2024 at 3:25 PM chuyuanchen ***@***.***> wrote:
Did you launch the code using mpirun or mpiexec?
Parallel {
Groups = 2
UseOmpThreads = Yes
}
$ mpirun -n 2 dftb+ | tee output
Suppose you have 10 kpts, this will let each process to solve 5 kpts in
parallel.
—
Reply to this email directly, view it on GitHub
<#1504 (comment)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/BFOC6BNV6XD22DVGUTMKGSTZP7GQ3AVCNFSM6AAAAABLI6N2OKVHI2DSMVQWIX3LMV43URDJONRXK43TNFXW4Q3PNVWWK3TUHMYTAMRUGY2DOMI>
.
You are receiving this because you authored the thread.Message ID:
***@***.***>
--
Best Regards
Er. Achyut Subedi
Registered Mechanical Engineer
(IOE, TU, Nepal)
+977-9847703110
|
Beta Was this translation helpful? Give feedback.
0 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
-
Hello everyone,
As suggested in manual, for parallel processing, I have been using 8 processes and my k point mesh is 101010 so in the groups section of parallel block, I entered 2 ( it should be integer divisor of process groups and k points) but I got the following error.
Processed input in HSD format written to 'dftb_pin.hsd'
Starting initialization...
STOP Error in cart_init: cannot build a 2D cartesian grid with incompatible sizes
ScriptError: You must enter a valid path to the input file.
When I change my groups to 1, the calculation works without error.
I am new to DFTB+ and I would be happy to get my problem resolved.
Thank you!
Beta Was this translation helpful? Give feedback.
All reactions