RAID-5, RAID-Z and RAID 1, RAID 10 compared

Gena Makhomed makhomed at pbank.lutsk.ua
Mon Sep 10 21:27:43 MSD 2007


Здравствуйте, Монашёв!

Monday, September 10, 2007, 7:47:51 PM, you wrote:

ММ> http://mail.opensolaris.org/pipermail/zfs-discuss/2007-September/042690.html

ММ> raidz оказался в 2 раза медленнее raid5, написанного pjd!

но RAID-5 имеет в принципе неустранимую "write hole" problem
кроме того, RAID-5 - достаточно сильно тормозит по сравнению
с RAID-10, если кроме чтения есть и запись. особенно на SATA

есть ли смысл искать новых приключений с RAID-5,
когда есть проверенные решения RAID-1 и RAID-10?

PS тормоза RAID-Z по сравнению с RAID-5
  - это плата за надежность и удобство.

=======================================================================

http://en.wikipedia.org/wiki/Standard_RAID_levels

RAID 5 "write hole" problem

In the event of a system failure while there are active writes,
the parity of a stripe may become inconsistent with the data.

If this is not detected and repaired before a disk or block fails,
data loss may ensue as incorrect parity will be used to reconstruct
the missing block in that stripe.

This potential vulnerability is sometimes known as the "write hole".

Battery-backed cache and similar techniques are commonly used
to reduce the window of opportunity for this to occur.

[...]

RAID-Z in the zfs filesystem of OpenSolaris
solves the "write hole" problem of RAID-5.

RAID-Z avoids the RAID 5 "write hole" by its copy-on-write policy:
rather than overwriting old data with new data,
it writes new data to a new location
and then atomically overwrites the pointer to the old data.

=======================================================================

-- 
Best regards,
 Gena                            mailto:makhomed at pbank.lutsk.ua







More information about the nginx-ru mailing list