Roman, did you create your zpool with ashift=12 on a 4k drive? You must manually ascertain whether your drives are 4k because they'll usually lie and be really slow.
On Tuesday, 4 March 2014, 11:11, Daniel Jozsef <daniel.joz...@gmail.com> wrote: I don't remember anyone ever saying that ZFS was fast. In fact it's a resource hog, and quite cumbersome. >It is flexible and safe. > > > >Your test is equivalent to saying that it's much easier to walk around with >money in a paper bag than in a reinforced steel security carrier case. > > > >(BTW, speaking of integrity, raid10 is nothing to write home about. It was a >surprise to me as well, but no raid configuration protects against silent data >corruption. The redundant raid configs were designed to protect from drive >failure.) > >On Sunday, October 27, 2013 6:50:49 PM UTC-4, Roman Kunz wrote: >I was playing around on Maverick with various FS setups. 4x 2TB drives in >raidz, raid10 (hfs+/zfs) and it seems raid10 hfs+ outperformes all other setup >by at least 80%. >>Zfs comp disabled, 128k blocks, only large data gets moved. Arc cache doesn't >>really kick in as most is write once / read once. Am I missing any tuneables? >>I don't really want to use hfs+ again but having on a simple stripe test 2x >>2TB read/write with zfs ~120MB/150MB i get with hfs ~200MB/250MB. >>-Roman -- > >--- >You received this message because you are subscribed to the Google Groups >"zfs-macos" group. >To unsubscribe from this group and stop receiving emails from it, send an >email to zfs-macos+unsubscr...@googlegroups.com. >For more options, visit https://groups.google.com/groups/opt_out. > > > -- --- You received this message because you are subscribed to the Google Groups "zfs-macos" group. To unsubscribe from this group and stop receiving emails from it, send an email to zfs-macos+unsubscr...@googlegroups.com. For more options, visit https://groups.google.com/groups/opt_out.
