ACC_CUFFT_MAKEPLAN with version 6.3.2 with GPU

questions related to VASP with GPU support (vasp.5.4.1, version released in Feb 2016)

Moderators: Global Moderator, Moderator

Post Reply
Message
Author
jsarlo@central.uh.edu
Newbie
Newbie
Posts: 4
Joined: Fri Sep 11, 2020 3:24 pm

ACC_CUFFT_MAKEPLAN with version 6.3.2 with GPU

#1 Post by jsarlo@central.uh.edu » Tue Sep 19, 2023 4:21 pm

We have built VASP 6.3.2 with NVHPC 22.5 to use with GPUs. The GPUs we have are V100 with 32G of memory. We are requesting 32 processors and 2 GPUs on the compute node. When the job runs, we get the following error

running on 32 total cores
distrk: each k-point on 32 cores, 1 groups
distr: one band on 1 cores, 32 groups
OpenACC runtime initialized ... 2 GPUs detected
WARNING: INIT_ACC: several MPI-ranks need to share a GPU, which is not
supported by NCCL. The use of NCCL will be switched off. To avoid this,
reduce the number of MPI-ranks: #-of-ranks <= #-of-GPUs (on every node!).

vasp.6.3.2 27Jun22 (build Dec 21 2022 14:13:29) complex
POSCAR found type information on POSCAR Pt
POSCAR found : 1 types and 55 ions
scaLAPACK will be used selectively (only on CPU)
-----------------------------------------------------------------------------
| |
| W W AA RRRRR N N II N N GGGG !!! |
| W W A A R R NN N II NN N G G !!! |
| W W A A R R N N N II N N N G !!! |
| W WW W AAAAAA RRRRR N N N II N N N G GGG ! |
| WW WW A A R R N NN II N NN G G |
| W W A A R R N N II N N GGGG !!! |
| |
| For optimal performance we recommend to set |
| NCORE = 2 up to number-of-cores-per-socket |
| NCORE specifies how many cores store one orbital (NPAR=cpu/NCORE). |
| This setting can greatly improve the performance of VASP for DFT. |
| The default, NCORE=1 might be grossly inefficient on modern |
| multi-core architectures or massively parallel machines. Do your |
| own testing! More info at wiki/index.php/NCORE |
| Unfortunately you need to use the default for GW and RPA |
| calculations (for HF NCORE is supported but not extensively tested |
| yet). |
| |
-----------------------------------------------------------------------------

LDA part: xc-table for Pade appr. of Perdew
POSCAR found type information on POSCAR Pt
POSCAR found : 1 types and 55 ions
POSCAR, INCAR and KPOINTS ok, starting setup
-----------------------------------------------------------------------------
| |
| EEEEEEE RRRRRR RRRRRR OOOOOOO RRRRRR ### ### ### |
| E R R R R O O R R ### ### ### |
| E R R R R O O R R ### ### ### |
| EEEEE RRRRRR RRRRRR O O RRRRRR # # # |
| E R R R R O O R R |
| E R R R R O O R R ### ### ### |
| EEEEEEE R R R R OOOOOOO R R ### ### ### |
| |
| ACC_CUFFT_MAKEPLAN: could not create plan |
| |
| ----> I REFUSE TO CONTINUE WITH THIS SICK JOB ... BYE!!! <---- |
| |
-----------------------------------------------------------------------------

I am guessing we don't have something set up right, but not sure what direction to go.

Jeff
UH IT - HPC

merzuk.kaltak
Administrator
Administrator
Posts: 261
Joined: Mon Sep 24, 2018 9:39 am

Re: ACC_CUFFT_MAKEPLAN with version 6.3.2 with GPU

#2 Post by merzuk.kaltak » Fri Nov 24, 2023 2:38 pm

Dear jsarlo,

sorry for the late reply, apparently this post slipped through the cracks.
Could you please upload makefile.include as well as all input files including the OUTCAR.

Post Reply