EPW Forum

Hi Wu, As the error suggests, the number of tasks "#SBATCH -n 8 " (8 in your case) should be the same as the number of k-point parallelization. Simply put for your case you should use, "mpiexec.hydra -n $SLURM_NPROCS <location of QE bin>/epw.x -nk $SLURM_NPROCS -in <input>" Notice -nk should be the ...

Hi Rijan,

In that case your nbndsub should be 40 and not 56.

Best,
Sabya.

Hi Rijan,

Please attach your files in order for us to debug. I am not sure what code is even reading in your case since you are supplying both filkf and nkf. You are supposed to supply either of them but not both.

Best,
Sabya.

Hi Rijan,

As the error says, you should define either "filkf" or nkf" and not both as you have in your input file.

Best,
Sabya.

Hi Rijan, What you have mentioned in the post and your epw1.in appear different. As you said you have 2 Hf atoms so 2*5(d)=10, and 8 Te atoms so 8*3(p)=24. So in total your nbndsub=34 if you are not performing a non-collinear calculation. Please check again. Otherwise, attach all your files so that ...

Hi Felix,

Thanks for pointing out. Indeed the link did not work. It has now been fixed.

Best regards,
Sabya.

Hi Liu and mxh, The system_2D in EPW 5.8 has been implemented using multiple formalisms. As already mentioned in the description of the input variables, I am mentioning it here again, the dipole_sp option will perform 2D truncation using "https://journals.aps.org/prb/abstract/10.1103/PhysRevB.107.15...

Hi Yao, Indeed, that is quite long but what I can see in your input is that you are using 128 bands. That is quite a lot!. Also, keep in mind that epw is only parallelized over the k-points. Therefore, since you have very few k-points (ignoring symmetries: 216) most of the processors are probably no...

Hi Albert,

This error has been fixed in EPW v5.8. Please download the developer version from "https://gitlab.com/epw/q-e.git".

Best regards,
Sabya.

Hi, From a first look it appears that your system where you are running EPW does not have enough memory. If it is an HPC, you can try to increase the number of nodes (while keeping the same number of cores) and see if this error goes away. Or reduce the nkf, nqf and check if it goes away. Otherwise,...