SOLVED how to fast move a big file to other datasets

[qq8554650]

all datasets in a disk
a_dataset\bigfile
b_dataset\

when I mv bigfile to b_dataset,like cp cost time
how to fast move it

 

[Ericloewe]

It you’re asking how to atomically move data between datasets, the answer is you don’t. Dataset boundaries are filesystem boundaries, subject to all the usual semantics.

 

[Arwen]

As @Ericloewe said, dataset boundries are file system boundries. This is on purpose because ZFS lets you change characteristics of one dataset and not another. Meaning you can have LZ4 compression on the source, and no compression on the destination. (Or change in checksum algorithms, etc...).

So, it's by design that moving a file, (large or small), is an actual copy & remove source process underneath.