| From: | "Jim Nasby" <jnasby(at)pervasive(dot)com> |
|---|---|
| To: | <pgsql-performance(at)postgresql(dot)org> |
| Subject: | FW: pg_xlog on data partition with BBU RAID |
| Date: | 2006-06-10 02:24:47 |
| Message-ID: | 4D27CB1096EF1C408F4BFAB0046EC7B608C241@ausmailid.aus.pervasive.com |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-performance |
Forwarding to -performance
From: Alan Hodgson [mailto:ahodgson(at)simkin(dot)ca]
On Friday 09 June 2006 12:41, "Jim C. Nasby" <jnasby(at)pervasive(dot)com> wrote:
> Has anyone actually done any testing on this? Specifically, I'm
> wondering if the benefit of adding 2 more drives to a RAID10 outweighs
> whatever penalties there are to having pg_xlog on that RAID10 with all
> the rest of the data.
I have an external array with 1GB of write-back cache, and testing on it
before deployment showed no difference under any workload I could generate
between having pg_xlog on a separate RAID-1 or having it share a RAID-10
with the default tablespace. I left it on the RAID-10, and it has been
fine there. We have a very write-heavy workload.
--
"If a nation expects to be ignorant and free, in a state of civilization,
it expects what never was and never will be." -- Thomas Jefferson
| From | Date | Subject | |
|---|---|---|---|
| Next Message | John Top-k apad | 2006-06-11 06:35:20 | Variation between query runtimes |
| Previous Message | Scott Marlowe | 2006-06-09 20:21:22 | Re: pg_xlog on data partition with BBU RAID |