Re: Maximum Possible Insert Performance?

From: Greg Stark <gsstark(at)mit(dot)edu>
To: William Yu <wyu(at)talisys(dot)com>
Cc: pgsql-performance(at)postgresql(dot)org
Subject: Re: Maximum Possible Insert Performance?
Date: 2003-11-25 01:16:27
Message-ID: 87brr11c44.fsf@stark.dyndns.tv
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-performance

William Yu <wyu(at)talisys(dot)com> writes:

> > You're right, though, mirroring a solid state drive is pretty pointless; if
> > power fails, both mirrors are dead.
>
> Actually no. Solid state memory is non-volatile. They retain data even without
> power.

Note that flash ram only has a finite number of write cycles before it fails.

On the other hand that might not be so bad for WAL which writes sequentially,
you can easily calculate how close you are to the maximum. For things like
heap storage or swap it's awful as you can get hot spots that get written to
thousands of times before the rest of the space is used.

--
greg

In response to

Responses

Browse pgsql-performance by date

  From Date Subject
Next Message Neil Conway 2003-11-25 02:24:45 Re: Optimize
Previous Message Christopher Kings-Lynne 2003-11-25 01:13:51 Re: Where to start for performance problem?