Re: Having I/O problems in simple virtualized environment

From: Claudio Freire <klaussfreire(at)gmail(dot)com>
To: Ron Arts <ron(dot)arts(at)gmail(dot)com>
Cc: pgsql-performance(at)postgresql(dot)org
Subject: Re: Having I/O problems in simple virtualized environment
Date: 2012-01-29 23:01:08
Message-ID: CAGTBQpZPDw3s12YSXL+tUzd0ckCgT47d7g+-_TR=thiLt_kp0Q@mail.gmail.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-performance

On Sun, Jan 29, 2012 at 7:48 PM, Ron Arts <ron(dot)arts(at)gmail(dot)com> wrote:
> Hi list,
>
> I am running PostgreSQL 8.1 (CentOS 5.7) on a VM on a single XCP (Xenserver) host.
> This is a HP server with 8GB, Dual Quad Core, and 2 SATA in RAID-1.
>
> The problem is: it's running very slow compared to running it on bare metal, and
> the VM is starving for I/O bandwidht, so other processes (slow to a crawl.
> This does not happen on bare metal.

My experience with xen and postgres, which we use for testing upgrades
before doing them on production servers, never in production per-se,
is that I/O is very costly on CPU cycles because of the necessary talk
between domU and dom0.

It's is worthwhile to pin at least one core for exclusive use of the
dom0, or at least only let low-load VMs use that core. That frees up
cycles on the dom0, which is the one handling all I/O.

You'll still have lousy I/O. But it will suck a little less.

In response to

Browse pgsql-performance by date

  From Date Subject
Next Message Carlo Stonebanks 2012-01-29 23:04:53 Re: pl/pgsql functions outperforming sql ones?
Previous Message Ron Arts 2012-01-29 22:48:52 Having I/O problems in simple virtualized environment