When using zstd (or anything else) make sure you don't have something in your .sss/config that would make ssh use a compression as you'll waste a bit of CPU to send a little more data as a less efficient compression method is applied on top. Though TBH I think that unless I'm on a very slow link, if I'm sending something big enough to need compression it is probably in a pre-compressed or otherwise uncompressable format anyway (video, photos, …) so zstd does little, the one exception I can think of being if I'm throwing an unencrypted VM image from place to place.
I never bothered to learn -C, instead I use:
For doing something like duplicating root of a filesystem you may want --one-file-system.with a sandwiched `| pv |` for fun stats
And `| ssh <target> ` for a remote copy.
(or ssh <target> prepended rather than sandwiched, to copy from remote)
At some point it starts very much looking like zfs send | pv | ssh <target> zfs receive
Seriously. I do wish ZFS was everywhere.
I usually sandwich zstd and base64 encode/decode in with the ssh pipe. Used when I need root permissions to pull the remote dirs.
You shouldn't need to B64 encode for transfer - piping over SSH is binary safe. Verifying with 100Mbyte of random data:
When using zstd (or anything else) make sure you don't have something in your .sss/config that would make ssh use a compression as you'll waste a bit of CPU to send a little more data as a less efficient compression method is applied on top. Though TBH I think that unless I'm on a very slow link, if I'm sending something big enough to need compression it is probably in a pre-compressed or otherwise uncompressable format anyway (video, photos, …) so zstd does little, the one exception I can think of being if I'm throwing an unencrypted VM image from place to place.