GistTree.Com
Entertainment at it's peak. The news is by your side.

SnapRAID, unRAID, ZFS and Btrfs comparison table

0

SnapRAID is easiest one in every of the readily accessible now not fashioned RAID solutions for disk arrays.

The finest known others are:

  • unRAID
    – Industrial and OpenSource GPL2 solution. A modified model of the Linux Reiserfs filesystem with proper-time redundancy.
    No integrity checksum is supported.
  • FlexRAID
    – Industrial and proprietary C++/Java application for Dwelling windows, with some shrimp give a steal to for Linux.
    It helps both snapshot redundancy and proper-time redundancy, with integrity checksum.
  • ZFS
    – OpenSource filesystem (but GPL incompatible) with proper-time redundancy and integrity checksum.
  • Btrfs
    – OpenSource filesystem GPL2 with proper-time redundancy.
    From Linux 3.9 it helps RAID5/6 redundancy and integrity checksums.
    Recently varied disorders had been discovered and the skilled wiki now states:
    The parity RAID code has plenty of serious data-loss bugs in it. It must nonetheless now not be feeble for something else varied than finding out functions.
  • Storage Spaces
    – The final contender from Microsoft, and constructed-in into Dwelling windows 8.
    Proprietary with proper-time redundancy. Checksumming is now not supported.
    It furthermore has some varied obstacles.

The essential aspect to categorize them is when the redundancy data is as a lot as this level.
In proper-time solutions the parity data is as a lot as this level in proper-time,
with out an explicit action from the actual person. Devour in fashioned RAID.
In snapshot solutions the parity data is as a lot as this level easiest upon
an explicit ask from the actual person. Devour in backups.

Other principal components are the give a steal to of an integrity checksum
to name restful data corruption
and the possibility to fix it,
if it is doubtless you’ll presumably perchance use already stuffed disk,
if it is doubtless you’ll presumably perchance discover better your data from now not broken disks if you discover too many failures to non-public a rotund discover better,
and if your total disks are spinning when reading one file.

SnapRAID unRAID FlexRAID ZFS Btrfs Storage Spaces
Redundancy model

Real-time computes parity at proper-time worship RAID,
snapshot at ask time, worship Backup.
Snapshot Real-time Real-time

Snapshot
Real-time

Snapshot
Real-time

Snapshot
Real-time
Integrity


If data is validated with some roughly checksum,
and which one is the default.
Poke

SpookyHash

128 bit
No [8] Poke

Adler32

32 bit
Poke

fletcher4

256 bit
Poke

CRC32C

32 bit
No [9]
Fix restful errors


If restful errors are detected and mounted prior to they’ll propagate into the parity.
Poke No [1] No [2] Poke [3] Poke [3] No
Number of failures


What number of disk failures are supported? 1 is for RAID5, 2 for RAID6.
1 2 3 4 5 6 1 2 1 2 3 4 5 6+ 1 2 3 1 2 1 2
Other failures

If more disks fail than the supported redundancy model,
is it imaginable to discover better the facts in the now not failed disks?
Poke Poke Poke No No No
Power/Noise

What number of disks are spinning when reading a single file?
One One One All All All
Stuffed disks

Are you able to originate with already stuffed disks?
Poke Partial[4] Poke No No No
Add disks

Are you able to add disks at later time?
Poke Poke Poke No[5] Poke Poke
Working machine

Which OS is supported?
Linux

Dwelling windows

Mac OS X

OpenIndiana

Solaris

BSD
Linux Dwelling windows

Linux
Linux [6]

Mac OS X

OpenIndiana

Solaris

BSD
Linux Dwelling windows
Age

The twelve months of the main skilled originate supporting a minimal of RAID5 redundancy.
2011 2005 2008 2005 now not yet stable [11] 2012
License/Sign


Tool license and tag.
Open Supply GPL3

Free
Open Supply GPL2

89$/129$
Proprietary

40$/60$
Open Supply CDDL

Free
Open Supply GPL2

Free
Proprietary

Dwelling windows 8/10
Interface

Which interface is supplied? GUI or expose line?
expose line

or Elucidate GUI,
or plugin for for OpenMediaVault
expose line

GUI
GUI expose line

GUI/Plugins [7]
expose line

GUI[10]
GUI

Notes

[1] – unRAID would now not non-public any roughly checksum, and it supreme ignores restful errors.
Even worse, if a parity error is detected as consequence of a restful error in the facts,
the parity is robotically recomputed, making now not doable to discover better the restful error, even manually.

[2] – Flexraid uses checksums to validate files, but such checksums
are now not verified when data is learn to update the parity.
This implies that any restful error latest will propagate into the parity,
making now not doable to fix it later, even supposing it goes to furthermore furthermore be nonetheless detected evaluating the file checksum.



You would furthermore discover in a order the keep aside the “Validate” operation
reviews errors, however the “Compare” one reviews no subject in the parity,
making now not doable to fix the errors.

[3] – ZFS and Btrfs provide a minute-rot safety on the identical level of SnapRAID,
continuously checking data prior to the utilization of it.
In this regards the total three solutions characterize the order-of-the-artwork.



A cons of ZFS is that the default Fletcher
checksum is a different that favorites tempo over quality.
The identical for the default CRC32C feeble by Btrfs.
The 128 bits SpookyHash
feeble by SnapRAID is as a change the order-of-the-artwork in checksumming quality,
with out compromising in tempo.



One other cons of ZFS is that it lacks a immediate RAID implementation in assembler.
It easiest has a C implementation, that is from two to four cases slower than SnapRAID/Btrfs.



ZFS furthermore uses a sub-optimum RAID-Z3 algorithm, that requires double computations than the equivalent SnapRAID’s z-parity.



As a change, both SnapRAID and Btrfs use high-notch assembler implementations to compute the RAID parity,
continuously the utilization of the finest known RAID algorithm and implementation.

[4] – unRAID permits to make use of stuffed disks but easiest in the occasion that they’re already
formatted with the ReiserFS, XFS or Btrfs filesystems.
However now not ext4 or NTFS, the 2 most trendy filesystems feeble in Linux and Dwelling windows.

[5] – ZFS permits to add a brand fresh vdev on the pool, but now not a disk to a vdev.
Device The ‘Hidden’ Payment of The use of ZFS for Your Dwelling NAS.

[6] – ZFS is furthermore readily accessible in Linux at kernel level the utilization of ZFS on Linux.

[7] – ZFS has external GUIs, worship Napp-it
and plugins for FreeNAS and NAS4Free.

[8] – unRAID can non-public integrity checksum the utilization of a plugin worship
Dynamix File Integrity,
Checksum Suite or
bunker
but they’re all honest on the parity processing and never feeble to assist the convalescing process.
For example, when convalescing a single failed disk with double parity,
the checksum could presumably furthermore furthermore be feeble to name further restful errors, and nonetheless have the selection to discover better.

[9] – Storage Spaces could presumably furthermore furthermore be feeble with ReFS that give a steal to integrity checksum, but now not in the Parity mode:
“Though, it is doubtless you’ll presumably perchance have the selection to buy Easy (no resiliency) or Parity, both of these alternatives will fail the technique”.

[10] – Btrfs has external GUIs, worship Rockstor.

[11] – Btrfs helps RAID5 from 2013, but now not too long in the past some serious disorders in the
RAID5/6 give a steal to had been discovered.
At latest they are now not mounted yet.
The skilled wiki says:
“The parity RAID code has plenty of serious data-loss bugs in it. It must nonetheless now not be feeble for something else varied than finding out functions”.

Need to you seen that some data reported is fallacious or simply unfair, please story it in the Forum.

Read More

Leave A Reply

Your email address will not be published.