Unwitting fragmentation on initial write?

[dashoe]

I recently rebuilt the pool on a server to optimize the configuration. Then to speed up the process of restoring the data, I copied from multiple sources (scp, Rsync, Hypervisor VDI migrate) simultaneously. Afterwords, I realized I may have unwittingly fragmented my data by writing multiple streams at the same time. Is the correct? Would I have been better off to copy the data from one source at a time? I was suprised that zpool list -v shows only 1% fragmentation. Perhaps TrueNas (or zfs) is smart about multiple write streams? This is on a 3 x RAIDZ2 | 7 wide | 931.51 GiB. Just curious about best practices.

Code:

NAME                                       SIZE  ALLOC   FREE  CKPOINT  EXPANDSZ   FRAG    CAP  DEDUP    HEALTH  ALTROOT
...
poolB                                     19.0T  3.38T  15.6T        -         -     1%    17%  1.00x    ONLINE  /mnt
...