Neko on Piz Daint

[njansson]

To compile Neko on Piz Daint

(For a general guide on how to build Neko for NVIDIA GPUs please see: #540)

• Make sure have to have the following modules loaded:
  • GPU software stack (module load daint-gpu)
  • Latest Cray Development Toolkit (module load cdt/22.05)
  • Cray Accelerator Target (module load craype-accel-nvidia60)
  • Third Party Scientific Libraries (module load cray-tpsl) (optional)
• Configure Neko to use CUDA, device aware MPI and ParMETIS (optional):

./configure  FC=ftn CC=cc --with-cuda --with-parmetis --enable-device-mpi

• Build using make && make install
• Make sure to set MPICH_RDMA_ENABLED_CUDA=1 in the shell/jobscript before running neko (otherwise device aware MPI will cause a segmentation fault), if everything went well backend type should be reported as CUDA, and the type of the device should be visible.

    _  __  ____  __ __  ____
   / |/ / / __/ / //_/ / __ \
  /    / / _/  / ,<   / /_/ /
 /_/|_/ /___/ /_/|_|  \____/

 (version: 0.4.99)
 (build: 2022-09-07 on x86_64-pc-linux-gnu using cray)


 -------Job Information--------
 Start time: 11:59 / 2022-09-07
 Running on: 2 MPI ranks
 Bcknd type: Accelerator (CUDA)
 Dev. name : Tesla P100-PCIE-16GB
 Real type : double precision