News for hackers. Stuff that matters.

A comment on In Praise of ZFS on Linux's ZED 'ZFS Event Daemon'. Parent comment

Re:In Praise of ZFS on Linux's ZED 'ZFS Event Daemon' by accelbred on Monday July 20, 2020 @04:25PM (#23901530)
If you're using RAID-Z on zfs, your comparison isn't fair. Rather than use RAID56 with btrfs, the equivalent would be to get 1 or 2 disk redundancy with raid1 or raid1c3.
[ Parent | on HN ]
Re:In Praise of ZFS on Linux's ZED 'ZFS Event Daemon' by tw04 on Tuesday July 21, 2020 @10:16AM (#23906935)
RAID-Z is the equivalent of RAID-5. RAID-Z2 is the equivalent of RAID-6. RAID-Z3 would be the equivalent of RAID-7 (or whatever the standard is named for 3-disk parity).

This is strictly speaking to how it deals with data and parity, the implementations are obviously different.

RAID-1 would be a mirror in ZFS parlance.

[ Parent | on HN ]
Re:In Praise of ZFS on Linux's ZED 'ZFS Event Daemon' by accelbred on Thursday July 23, 2020 @02:26AM (#23924706)
BTRFS raid1 isn't mirroring drives though, it means there are two copies of each extent across the whole set of 2+ drives. and BTRFS and raid1c3 and raid1c4 are 3 and 4 copies.
[ Parent | on HN ]
Re:In Praise of ZFS on Linux's ZED 'ZFS Event Daemon' by trasz on Monday July 20, 2020 @08:19PM (#23903226)
Not really - RAIDZ is basically RAID5 without the write hole problem. ZFS equivalent to RAID1 is called 'mirror', and is... well, a mirror.
[ Parent | on HN ]
Re:In Praise of ZFS on Linux's ZED 'ZFS Event Daemon' by accelbred on Thursday July 23, 2020 @02:28AM (#23924719)
Yes, and btrfs's raid1 isn't a mirror. Btrfs's raid1c3 and raid1c4 are its alternative to raid5/6 without the write hole.
[ Parent | on HN ]
Re:In Praise of ZFS on Linux's ZED 'ZFS Event Daemon' by trasz on Friday July 24, 2020 @06:02PM (#23944947)
No, it's still a mirror, just spread over more than two devices. Size overhead is still that of a mirror, ie quite a bit higher than that of raidz.
[ Parent | on HN ]