Re: suggestions for postgresql setup on Dell 2950 , PERC6i controller

Sorry for the top post --

Assuming Linux --

1: PERC 6 is still a bit inferior to other options, but not that bad. Its random IOPS is fine, sequential speeds are noticeably less than say the latest from Adaptec or Areca.

2: Random iops will probably scale ok from 6 to 8 drives, but depending on your use case, having a single mirror for OS and xlogs can be a significant performance improvement. My base suggestion would be to go with 6 drives in raid 10, perhaps try 3 mirrors with software raid 0 on top to compare, and leave one mirror for the OS and xlogs (separate partitions, use ext2 for the xlogs and that can be a fairly small partition). There isn't any way to know which of these options will be best for you, it is very dependant on the data and applications accessing it.

3. No, its too hardware dependant to have an ideal raid block size. Its slightly usage dependant too. The default is probably going to be best based on my PERC 6 experience. You'll gain a lot more from tuning other things.

4. Can't say much about the SAN. High end ones can do good iops, the one listed looks more like archival storage to me though.

Make sure you tune the Linux block device readahead, it makes a huge difference in sequential access performance ( see blockdev -getra <device>). 1MB to 4MB per raid spindle 'width' is usually ideal. The default is 128k, the default with software raid is 4MB, often the performance difference when using software raid is largely just this setting. If you're comfortable with XFS, it works well for the postgres data files.

On 2/4/09 10:45 AM, "Rajesh Kumar Mallah" <mallah(dot)rajesh(at)gmail(dot)com> wrote:

Hi,

I am going to get a Dell 2950 with PERC6i with
8 * 73 15K SAS drives +
300 GB EMC SATA SAN STORAGE,

I seek suggestions from users sharing their experience with
similar hardware if any. I have following specific concerns.

1. On list i read that RAID10 function in PERC5 is not really
striping but spanning and does not give performance boost
is it still true in case of PERC6i ?

2. I am planning for RAID10 array of 8 drives for entrire database
( including pg_xlog) , the controller has a write back cache (256MB)
is it a good idea ?
or is it better to have 6 drives in HW RAID1 and RAID0 of 3 mirrors
in s/w and leave 2 drives (raid1) for OS ?

3. Is there any preferred Stripe Size for RAID0 for postgresql usage ?

4. Although i would benchmark (with bonnie++) how would the EMC
SATA SAN storage compare with locally attached SAS storage for the
purpose of hosting the data , i am hiring the storage primarily for
storing base base backups and log archives for PITR implementation.
as retal of separate machine was higher than SATA SAN.

Regds
mallah.

--
Sent via pgsql-performance mailing list (pgsql-performance(at)postgresql(dot)org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-performance