Author Topic: mobility_error  (Read 92 times)

0 Members and 1 Guest are viewing this topic.

Offline Roc2019

  • Regular ATK user
  • **
  • Posts: 35
  • Country: cn
  • Reputation: 0
    • View Profile
mobility_error
« on: July 8, 2020, 04:31 »
Dear Sir,

I can follow the mobility examples of graphene and get the same result.  But, as I did a test to calculate the mobility of graphene by the method of spin + U,  the following error is printed.  So,  what's the matter?   Thank you.  The attachment is the input file of the spin + U calculation.
...........
..........
ElectronPhononCoupling: Done with 80 % of the calculations. ETC 1:12:25
ElectronPhononCoupling: Done with 90 % of the calculations. ETC 0:36:10
Traceback (most recent call last):
  File "input.py", line 276, in <module>
    use_kpoint_symmetry=True
  File "zipdir/NL/ComputerScienceUtilities/Functions.py", line 520, in method_wrapper
  File "zipdir/NL/Analysis/ElectronPhononCoupling.py", line 428, in __init__
  File "zipdir/NL/Analysis/ElectronPhononCoupling.py", line 1978, in calculateElectronPhononCoupling
  File "zipdir/NL/Analysis/ElectronPhononCoupling.py", line 2718, in _performElectroPhononCouplingCalculation
  File "zipdir/NL/ComputerScienceUtilities/ParallelTools/ParallelTools.py", line 494, in sumAcrossProcesses
  File "zipdir/NL/ComputerScienceUtilities/ParallelTools/ParallelTools.py", line 528, in _sumNumpyArrayAcrossProcesses
  File "Comm.pyx", line 545, in mpi4py.MPI.Comm.Allreduce (src/mpi4py.MPI.c:68568)
mpi4py.MPI.Exception: Invalid count, error stack:
MPI_Allreduce(1562): MPI_Allreduce(sbuf=0x2ad1e5f60010, rbuf=0x2add2cb18010, count=-1292891296, dtype=0x4c001041, MPI_SUM, MPI_COMM_WORLD) failed
MPI_Allreduce(1500): Negative count, value is -1292891296
« Last Edit: July 8, 2020, 04:40 by Roc2019 »

Offline Troels-Markussen

  • Regular ATK user
  • **
  • Posts: 19
  • Country: dk
  • Reputation: 1
  • QuantumATK staff
    • View Profile
Re: mobility_error
« Reply #1 on: July 9, 2020, 14:14 »
Hi,

This looks strange and should not happen. How many processes did you run with?  I was unable to reproduce the error, with a slightly modified script. Perhaps try to run with fewer processes. Also, I noticed that your k-point sampling for the ElectronPhononCoupling is around the Gamma-point (ka_range=[-0.1, 0.1], kb_range=[-0.1, 0.1]), while I guess you would rather have it around the K-point, i.e.

Code: [Select]
kpoints = RegularKpointGrid(
    ka_range=[0.23, 0.43],
    kb_range=[0.23, 0.43],
    kc_range=[0.0, 0.0],
    na=20,
    nb=20,
    nc=1,
    )

Regards,
Troels

Offline Roc2019

  • Regular ATK user
  • **
  • Posts: 35
  • Country: cn
  • Reputation: 0
    • View Profile
Re: mobility_error
« Reply #2 on: July 10, 2020, 04:20 »
Thank you Dr. Markussen .

But,  how to obtain the mobility for spin up  and spin down states, respectively?

Thanks again.
« Last Edit: July 10, 2020, 05:12 by Roc2019 »