Just curious, on the host side, isn't this also an issue with how the disk label is created and how many bytes it consumes? If the label is not an 8k multiple, like an EFI label used by default when you give ZFS a whole lun, then wouldn't the ZFS data start at an unaligned boundary?
In the final article of this
series, I walk through the fascinating topic of block picking.
As is well known ZFS is an allocate on write...
In the initial days of ZFS some pointed out that ZFS resilvering
was metadata driven and was therefore super fast : after all we
only had to...
Today, we're making performance headlines with Oracle's ZFS Storage
SPC-1 : Twice the performance of NetApp at the same latency;...