Bienvenido! - Willkommen! - Welcome!

Bitácora Técnica de Tux&Cía., Santa Cruz de la Sierra, BO
Bitácora Central: Tux&Cía.
Bitácora de Información Avanzada: Tux&Cía.-Información
May the source be with you!

Wednesday, October 20, 2010

ZFS vs. Linux Raid vs. Linux LVM vs. Linux LVM + Raid

Source By James Dickens
uadmin.blogspot.com
Submit changes, enhancements, or comments to jamesd DOT wi AT gmail DOT com


ZFS
Linux 
Raid
Linux
 LVM
Linux 
LVM
+
Raid
Raid Types




Single drive
Yes
N/A
Yes
Yes
Raid-0
Yes
Yes
Yes
Yes
Raid-1
Yes
Yes
No
Yes
Raid-1+0
Yes
No
No
Yes
Raid-5
Noi
Yes
No
Yes
Raid-5+0
No
No
No
Yes
Raidzii
Yes
No
No
No
Raidz+0
Yes
No
No
No





Administrative tasks




Non-endian 
specific
Yes
No
No
No
Import/Export Pools/Dataset
Yes
Not automated
Not 
automated
Not automated
iostat built to
 gives details 
of IO utilization
Yes
No
No
No





Snapshots




Snapshots/online?
Yes/online
No
Yes/offline only
Yes/offline only
Requires separate slice
No
N/A
Yes
Yes
Uses more
 than 1% 
of data space
 to create
No
N/A
Yes
Yes
Rollbackiii
Yes
No
No
No
Clonesiv
Yes
No
No
No





Growing File Systems Online




Convert from
 single drive
 to raid1
Yes
Yes
No
Yes
Convert from Raid1 to Raid 1+0
Yes
N/A
N/A
Yes
Convert from Raid1 to Raid5/z
No/may be
 possible 
in the future
Yes/offline
N/A
Yes/offline
Grow a raid0
Yes/Yes
Yes/Yes
Yes/Yes
Yes/yes
Grow a raid5/z
No/may be
 possible
in the future
Yes/offline
Yes/offline
Yes/offline
Convert from Raid5/z to Raid5/z+0
Yes/Yes
No
No
Yes/yes





Coping with difficulties




Handles whole 
disk failure
Yes
Yes
Yes
Yes
Checksum's
 all data
Yes
No
No
No
Copes with
 disk corruption
Yes
Nov
Nov
Nov





Special Attributes




Optional On Disk Compression
Yes built in.
No/requires
 another layer
 of software
No/requires
 another layer
 of software
No/requires
 another 
layer of software
Multiple volume/filesystems per device/pool
Yes
No
Yes
Yes
Integrated NFS support
Yes
No
No
No
Recursively 
apply File 
system attributes
Yes
No
No
No
Built-in backup/restore support
Yes
No
No
No
Space Shared between all filesystems
 in the
 pool/dataset
Yes
One filesystem 
per device
You preallocate
 space per 
filesystem
You preallocate space 
per filesystem
Integrated quotas
Per filesystem
Not Integrated
 but more
 flexible
Not Integrated
but more
 flexible
Not Integrated
 but more 
flexible
Integrated Reservations
Yes
No
No
No


Sources
  1. Solaris Express man pages, zpool(1M), zfs(1M)
Footnotes
i   ZFS doesn't support raid 5 but does support raid-z that has better features and less limitations
ii  RaidZ - A variation on RAID-5 which allows for better distribution of parity and eliminates the "RAID-5 write hole" (in which data and parity become inconsistent after a power loss). Data and parity is striped across all disks within a raidz group. A raidz group with N disks of size X can hold approximately (N-1)*Xbytes and can withstand one device failing before data integrity is compromised. The minimum number of devices in a raidz group is 2; the recommended number is between 3 and 9.
iii  Rollback the given dataset to a previous snapshot. When a dataset is rolled back, all data that has changed since the snapshot is discarded, and the dataset reverts to the state at the time of the snapshot.
iv  A clone is a writable volume or file system whose initialcontents are the same as another dataset. As with snapshots, creating a clone is nearly instantaneous, and initially consumes no additional space.
v  [Linux] RAID (be it hardware- or software-), assumes that if a write to a disk doesn't return an error, then the write was successful. Therefore, if your disk corrupts data without returning an error, your data will become corrupted. This is of course very unlikely to happen, but it is possible, and it would result in a corrupt filesystem. http://www.tldp.org/HOWTO/Software-RAID-HOWTO-6.html
ZFS related links:
Why ZFS for Home Moving ZFS pools ZFS root is near Zones on ufs vs. ZFS

No comments: