Docker containers are usually treated as immutable once they’ve started running. You can update some configuration parameters dynamically though, such as the container’s name and its hardware resource limits.
Docker lets you package up software as self-sufficient containers that share their host’s kernel but look and feel like independent VMs. The software within runs with near-native performance. This makes Docker containers a great way to safely try out new packages without installing them “bare metal” on your machine. Here’s some of the techniques you can use.
Docker containers don’t see your system’s GPU automatically. This causes reduced performance in GPU-dependent workloads such as machine learning frameworks. Here’s how to expose your host’s NVIDIA GPU to your containers.