[oi-dev] Improving file transfer speed on ZFS

Stephan Althaus Stephan.Althaus at Duedinghausen.eu
Tue Nov 4 07:06:43 UTC 2025


On 11/4/25 05:03, Atiq Rahman wrote:
> Hi,
> When I am copying over files to the zfs dataset on an external HDD, 
> the speed is one third of what it is over exFAT. This is zfs to zfs 
> copy which is basically way slower than zfs to exFAT.
>
> This was my zpool create command on the external drive,
>
> ```
> |zpool create -o ashift=12 -O compression=lz4 -o autotrim=off -O 
> mountpoint=none -O atime=off -O dnodesize=auto hyperspace /dev/dsk/c4t0d0|
> |```|
> |
> |
> |And, this is how the dataset was created:|
> |
> |
> |```|
> |
> |zfs create -o recordsize=1M -o normalization=formD -O 
> mountpoint=/media hyperspace/media|
> |```|
> |
>
> I know zfs has journaling overhead. But, is it possible to set certain 
> zfs properties to improve the transfer speed?
>
>
> Best!
>
> Atiq
>
>
> _______________________________________________
> oi-dev mailing list
> oi-dev at openindiana.org
> https://openindiana.org/mailman/listinfo/oi-dev

Hello Atiq!

How do you copy, please show the commands (?)

You may want to investigate the read speed with something like zfs send 
... > /dev/null

If you do a zfs send -> receive, a buffer like "mbuffer" may help, over 
network pathes in combination with a compression pbzip2, pigz ...


Regards,

Stephan

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://openindiana.org/pipermail/oi-dev/attachments/20251104/d9524274/attachment.html>


More information about the oi-dev mailing list