| From: | Craig Ringer <ringerc(at)ringerc(dot)id(dot)au> |
|---|---|
| To: | John R Pierce <pierce(at)hogranch(dot)com> |
| Cc: | pgsql-general(at)postgresql(dot)org |
| Subject: | Re: database corruption questions |
| Date: | 2012-10-14 03:26:40 |
| Message-ID: | 507A30F0.3040908@ringerc.id.au |
| Views: | Whole Thread | Raw Message | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-general |
On 10/14/2012 11:00 AM, John R Pierce wrote:
> On 10/13/12 7:13 PM, Craig Ringer wrote:
>>
>> * Use a good quality hardware RAID controller with a battery backup
>> cache unit if you're using spinning disks in RAID. This is as much for
>> performance as reliability; a BBU will make an immense difference to
>> database performance.
>
> a comment on this one.... I have some test servers with lots of SAS
> and/or SATA drives on controllers like LSI Logic 9261-8i, with 512MB or
> 1GB battery-backed cache. I can configure the controller for JBOD
> and use linux mdraid raid10 and get the same performance as the
> controllers native raid10, as long as the write-back cache is
> enabled. disable the writeback cache, and you might as well be using
> SATA JBOD.
Yeah, without the write-back cache you don't gain much. I run a couple
of DBs on plain old `md` RAID and I'm actually quite happy with it.
I've expanded this into a blog post and improved that section there.
http://blog.ringerc.id.au/2012/10/avoiding-postgresql-database-corruption.html
Comments appreciated.
--
Craig Ringer
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Craig Ringer | 2012-10-14 03:28:02 | Re: database corruption questions |
| Previous Message | John R Pierce | 2012-10-14 03:00:36 | Re: database corruption questions |