Re: Performance on 8CPU's and 32GB of RAM

From: "Scott Marlowe" <scott(dot)marlowe(at)gmail(dot)com>
To: "James Mansion" <james(at)mansionfamily(dot)plus(dot)com>
Cc: "Trevor Talbot" <quension(at)gmail(dot)com>, pgsql-performance(at)postgresql(dot)org
Subject: Re: Performance on 8CPU's and 32GB of RAM
Date: 2007-09-06 20:59:34
Message-ID: dcc563d10709061359y28b4098es786599bc49ae3c33@mail.gmail.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-performance

On 9/6/07, James Mansion <james(at)mansionfamily(dot)plus(dot)com> wrote:
> Scott Marlowe wrote:
> > And there's the issue that with windows / NTFS that when one process
> > opens a file for read, it locks it for all other users. This means
> > that things like virus scanners can cause odd, unpredictable failures
> > of your database.
> >
> >
> Can you provide some justification for this?

Seeing as I didn't write Windows or any of the plethora of anti-virus
software, no I really can't. It's unforgivable behaviour.

Can I provide evidence that it happens? Just read the archives of
this list for the evidence. I've seen it often enough to know that
most anti-virus software seems to open files in exclusive mode and
cause problems for postgresql, among other apps.

> Why do you think that UNIX systems are better at handling large shared
> buffers than Wndows?

Because we've seen lots of problems with large shared buffers on windows here.

Now, maybe for a windows specific app it's all fine and dandy. but
for the way pgsql works, windows and large shared buffers don't seem
to get along.

I'm done. Use windows all you want. I'll stick to unix. It seems to
just work for pgsql.

In response to

Browse pgsql-performance by date

  From Date Subject
Next Message Alvaro Herrera 2007-09-06 21:21:16 Re: Performance on 8CPU's and 32GB of RAM
Previous Message James Mansion 2007-09-06 20:55:11 Re: Performance on 8CPU's and 32GB of RAM