Re: Compared MS SQL 2000 to Postgresql 9.0 on Windows

From: Scott Marlowe <scott(dot)marlowe(at)gmail(dot)com>
To: Gael Le Mignot <gael(at)pilotsystems(dot)net>
Cc: Craig James <craig_james(at)emolecules(dot)com>, pgsql-performance(at)postgresql(dot)org
Subject: Re: Compared MS SQL 2000 to Postgresql 9.0 on Windows
Date: 2010-12-18 11:31:38
Message-ID: AANLkTimaOfpMmGKcbwWSfspQZ75L-tGw5jVuS5uGQA-z@mail.gmail.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-performance

2010/12/18 Gael Le Mignot <gael(at)pilotsystems(dot)net>:
> Hello Scott!
>
> Fri, 17 Dec 2010 19:06:15 -0700, you wrote:
>
>  > On Fri, Dec 17, 2010 at 10:32 AM, Craig James
>  > <craig_james(at)emolecules(dot)com> wrote:
>  >> RAID5 is a Really Bad Idea for any database.  It is S...L...O...W.  It does
>  >> NOT give better redundancy and security; RAID 10 with a battery-backed RAID
>  >> controller card is massively better for performance and just as good for
>  >> redundancy and security.
>
>  > The real performance problem with RAID 5 won't show up until a drive
>  > dies and it starts rebuilding
>
> I don't  agree with that. RAID5 is  very slow for random  writes, since
> it needs to :

Trust me I'm well aware of how bad RAID 5 is for write performance.
But as bad as that is, when the array is degraded it's 100 times
worse. For a lot of workloads, the meh-grade performance of a working
RAID-5 is ok. "Not a lot of write" data warehousing often runs just
fine on RAID-5. Until the array degrades. Then it's much much slower
than even a single drive would be.

In response to

Browse pgsql-performance by date

  From Date Subject
Next Message Pierre C 2010-12-18 13:55:26 Re: Compared MS SQL 2000 to Postgresql 9.0 on Windows
Previous Message selvi88 2010-12-18 09:34:42 Re: postgres performance tunning