Currently trying ZFS on my Linux machines. The experience is alright so far. Only problem is that ZFS on Linux is implemented as a kernel module, so I got to recompile ZFS each time the kernel is upgraded.
[[email protected] ~]# zpool list
NAME SIZE ALLOC FREE CAP DEDUP HEALTH ALTROOT
datapool 496G 5.17G 491G 1% 1.17x ONLINE –
This machine is a syslog server. I’m particularly curious about the de-duplication. This is a block level duplication, and there are a large amount of common stuff in syslog message so I guess the de-dup would be useful. I’m still trying basic stuff here though. Not too confident to use fanciful stuff on experimental software.
More information here – http://zfsonlinux.org/