Re: High end server and storage for a PostgreSQL OLTP system

From: Cosimo Streppone <cosimo(at)streppone(dot)it>
To: "Jim C(dot) Nasby" <decibel(at)decibel(dot)org>
Cc: Postgresql Performance list <pgsql-performance(at)postgresql(dot)org>
Subject: Re: High end server and storage for a PostgreSQL OLTP system
Date: 2005-02-01 06:35:35
Message-ID: 41FF2337.7070809@streppone.it
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-performance

Jim C. Nasby wrote:

> On Mon, Jan 31, 2005 at 09:41:32PM +0100, Cosimo wrote:
>
> >2) The goal is to make the db handle 100 tps (something like
> > 100 users). What kind of server and storage should I provide?
>
> You might look at Opteron's, which theoretically have a higher data
> bandwidth. If you're doing anything data intensive, like a sort in
> memory, this could make a difference.

Would Opteron systems need 64-bit postgresql (and os, gcc, ...)
build to have that advantage?

> >4) Is it correct to suppose that multiple RAID 1 arrays
> > can provide the fastest I/O ?
> > I usually reserve one RAID1 array to db data directory,
> > one RAID1 array to pg_xlog directory and one RAID1 array
> > for os and application needs.
>
> RAID10 will be faster than RAID1.

Sorry Jim, by RAID10 you mean several raid1 arrays mounted on
different linux partitions? Or several raid1 arrays that
build up a raid0 array? In the latter case, who decides which
data goes in which raid1 array? Raid Adapter?

> The key factor to a high performance database is a high
> performance I/O system. If you look in the archives
> you'll find people running postgresql on 30 and 40
> drive arrays.

I'll do a search, thank you.

--
Cosimo

In response to

Responses

Browse pgsql-performance by date

  From Date Subject
Next Message Richard Huxton 2005-02-01 08:50:54 Re: Very important choice
Previous Message Cosimo Streppone 2005-02-01 06:26:53 Re: High end server and storage for a PostgreSQL OLTP system