New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

cuda10 2080Ti mutil_gpu solutions #301

Open

zyxcambridge opened this issue Jul 24, 2019 · 0 comments

zyxcambridge commented Jul 24, 2019 •

edited

horovod
HOROVOD_GPU_ALLREDUCE=NCCL pip install --no-cache-dir horovod
Open MPI
3.0.4 no 2.1.1
wget http://www.open-mpi.org/software/ompi/v2.1/downloads/openmpi-2.1.1.tar.gz
tar xzvf openmpi-2.1.1.tar.gz
WORKDIR openmpi-2.1.1
./configure
make -j48
sudo make install
sudo ldconfig
which mpicc
mpicc -show
which mpiexec
mpiexec --version
mkdir OpenMpi
wget http://www.open-mpi.org/papers/workshop-2006/hello.c
mpicc hello.c -o hello
mpirun --allow-run-as-root -np 4 ./hello
3.eno1
options="-np ${nb_gpus} -H localhost:${nb_gpus} -bind-to none -map-by slot
-x NCCL_DEBUG=INFO -x NCCL_SOCKET_IFNAME=eno1 -x NCCL_IB_DISABLE=1
-x LD_LIBRARY_PATH --mca btl_tcp_if_include eth1

4.gcc 4.9 g++ 4.9 ;others not ok

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment