Re: Slow queries / commits, mis-configuration or hardware issues? - Mailing list pgsql-performance

From Tomas Vondra
Subject Re: Slow queries / commits, mis-configuration or hardware issues?
Date
Msg-id c22ff328a0f3e8d1ddd013aa8913dd09.squirrel@sq.gransy.com
Whole thread Raw
In response to Re: Slow queries / commits, mis-configuration or hardware issues?  (Cody Caughlan <toolbag@gmail.com>)
List pgsql-performance
On 16 Listopad 2011, 2:21, Cody Caughlan wrote:
> How did you build your RAID array? Maybe I have a fundamental flaw /
> misconfiguration. I am doing it via:
>
> $ yes | mdadm --create /dev/md0 --level=10 -c256 --raid-devices=4
> /dev/xvdb /dev/xvdc /dev/xvdd /dev/xvde
> $ pvcreate /dev/md0
> $ vgcreate lvm-raid10 /dev/md0
> $ lvcreate -l 215021 lvm-raid10 -n lvm0
> $ blockdev --setra 65536 /dev/lvm-raid10/lvm0
> $ mkfs.xfs -f /dev/lvm-raid10/lvm0
> $ mkdir -p /data && mount -t xfs -o noatime /dev/lvm-raid10/lvm0 /data
>
> --

I don't think you have a flaw there. The workload probably skews the
results a bit on the master and slave, so it's difficult to compare it to
results from an idle instance. The amount of data written seems small, but
a random i/o can saturated the devices quite easily.

I went with a very simple raid config - no LVM, default stripe size
(better seeks, worse sequential performance),  default read-ahead (could
give better seq. performance).

Tomas



pgsql-performance by date:

Previous
From: Cody Caughlan
Date:
Subject: Re: Slow queries / commits, mis-configuration or hardware issues?
Next
From: Greg Smith
Date:
Subject: Re: What's the state of postgresql on ext4 now?