Skip site navigation (1) Skip section navigation (2)


From: Bruce Momjian <bruce(at)momjian(dot)us>
To: Dan Langille <dan(at)langille(dot)org>
Cc: Matthew Wakeling <matthew(at)flymine(dot)org>, Greg Smith <greg(at)2ndquadrant(dot)com>, Laszlo Nagy <gandalf(at)shopzeus(dot)com>, pgsql-performance <pgsql-performance(at)postgresql(dot)org>
Subject: Re: SSD + RAID
Date: 2010-02-20 23:19:03
Message-ID: (view raw, whole thread or download thread mbox)
Lists: pgsql-performance
Dan Langille wrote:
> Hash: SHA1
> Bruce Momjian wrote:
> > Matthew Wakeling wrote:
> >> On Fri, 13 Nov 2009, Greg Smith wrote:
> >>> In order for a drive to work reliably for database use such as for 
> >>> PostgreSQL, it cannot have a volatile write cache.  You either need a write 
> >>> cache with a battery backup (and a UPS doesn't count), or to turn the cache 
> >>> off.  The SSD performance figures you've been looking at are with the drive's 
> >>> write cache turned on, which means they're completely fictitious and 
> >>> exaggerated upwards for your purposes.  In the real world, that will result 
> >>> in database corruption after a crash one day.
> >> Seagate are claiming to be on the ball with this one.
> >>
> >>
> > 
> > I have updated our documentation to mention that even SSD drives often
> > have volatile write-back caches.  Patch attached and applied.
> Hmmm.  That got me thinking: consider ZFS and HDD with volatile cache.
> Do the characteristics of ZFS avoid this issue entirely?

No, I don't think so.  ZFS only avoids partial page writes.  ZFS still
assumes something sent to the drive is permanent or it would have no way
to operate.

  Bruce Momjian  <bruce(at)momjian(dot)us>
  PG East:
  + If your life is a hard drive, Christ can be your backup. +

In response to


pgsql-performance by date

Next:From: George SextonDate: 2010-02-21 01:03:19
Subject: AutoVacuum_NapTime
Previous:From: Dan LangilleDate: 2010-02-20 23:07:16
Subject: Re: SSD + RAID

Privacy Policy | About PostgreSQL
Copyright © 1996-2017 The PostgreSQL Global Development Group