> The LOCALIO protocol support allows the NFS client and server to reliably determine if they are on the same host. If they happen to be on the same host, the network RPC protocol for read/write/commit operations are bypassed. Due to bypassing the XDR and RPC for reads/writes/commits, there can be big performance gains to using the LOCALIO protocol.
> One of the intended scenarios where it can be common for having the NFS client and server on the same host is when running containerized workloads
What are the advantages to NFS [with localio] over OverlayFS/overlay2 and bind mounts across context boundaries for container filesystems?
Do the new faster direct read() and write() IO calls have different risks compared to disk io with network IO?
Are the file permissions the same with localio? How does that compare to subuids and subuids for rootless containers?
I can imagine say a media server running NFS that can be accessed by both local containers and remote hosts.
What I think would be interesting is if you could access NFS on a local machine from a container or VM without using a network. You could have a standalone VM with no networking that could access shared files. (I think some people use 9p for this?)
When I saw it written as Localio I thought it was Italian !