Tom,
your hint was 100% correct. BINGO!
The setting of lc_locale and lc_ctype was 'en_us' instead of 'C' as
on the other machine.
Now it works perfectly OK and with identical execution times on both
machines.
Thanks for your help and best regards,
Rudolf VanderLeeden
Am 04.06.2008 um 16:45
Rudolf van der Leeden <[EMAIL PROTECTED]> writes:
> [ different times to build a varchar index on allegedly identical
> installations ]
Perhaps one is using C locale and the other is not? strcmp() vs
strcoll() is a pretty big hit. In general it'd be worth dumping
out the whole contents of pg_s
On Wed, Jun 04, 2008 at 11:43:24AM +0200, Rudolf van der Leeden wrote:
> I'm currently testing a 25G Postgres 8.3.0 database on Apple Xserve/
> Intel with Mac OSX 10.5 (Leopard).
>
> There are two identical machines with identical configurations (hw
> and sw).
> Running a pg_restore of a binar
Hi,
I'm currently testing a 25G Postgres 8.3.0 database on Apple Xserve/
Intel with Mac OSX 10.5 (Leopard).
There are two identical machines with identical configurations (hw
and sw).
Running a pg_restore of a binary backup file (3.8 GB) on both
machines gives the following results:
Ma
