![]() not possible to use with docker-compose and other tools for managing Docker containers.less transparent what is being executed (some believe some "black magic" happens behind nvidia-docker that handles 2 instances on same GPU better, although it works exactly the same). ![]() yet another tool that administrators need to learn (why bother administrators to learn anything more than docker run?).shorter command (no need to remember those additional parameters).Latest Docker introduced runtimes and nvidia-container-runtime should make things work, but unfortunately this is not supported by docker-compose and other tools. In reality the nvidia-docker is nothing more than a fancy wrapper that runs the docker command with the additional parameters to mount the device and host libraries into the container. ![]() It is true, that Nvidia recommends to use their nvidia-docker command as part of their vendor lock-in strategy. inject the correct version of CUDA Driver into the container as mentioned above (if it is installed on the host).upgrade your Nvidia kernel driver on the host by compiling it yourself.upgrade your Nvidia kernel driver on the host directly from Debian 9 packages: nvidia-kernel-dkms, nvidia-alternative, libnvidia-ml1, nvidia-smi.In case your NVIDIA Kernel Driver and CUDA Driver versions differ an error appears in kernel messages ( dmesg) or using nvidia-smi inside the container. $ watch -n 5 'nvidia-smi echo sensors for hdd in /dev/sd? do echo -n "$hdd " smartctl -A $hdd | grep Temperature_Celsius done ' To utilize your GPUs this Docker image needs access to your /dev/nvidia* devices and also the correct version of CUDA Driver, like: optionally nvidia-smi, nvidia-opencl-icd.CUDA Driver library ( libcuda1, same version as NVIDIA Kernel Driver).NVIDIA Kernel Driver ( nvidia-kernel-dkms). ![]()
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |