Hello Norbert,
I don't have the exact environment that we used for testing. However on
our production machines we have PostgreSQL v8.1.8 and the following
settings:
effective_cache_size = 4000
shared_buffers = 1000
Make sure you are communicating with postgres over the unix domain
socket and not via tcp/ip. After importing your data make sure you do a
vacuum analyze. Get postgres or powerdns to dump the sql statements and
run an explain within postgres to make sure your indexes are set up aok.
Norbert Sendetzky wrote:
Hi Adam
We tested this on Debian 'testing'. Pdns version is 2.9.20. In testing I
used the gmysql and gpgsql backends.
Postgresql is version 8.1 and Mysql version 5.0.32
I've tried to reproduce your test results but didn't succeeded. The test
environment is the same (PDNS 2.9.20, PostgreSQL 8.1) but with a much slower
machine (Via C3 533MHz) and only 256 MB RAM.
My test results are (PDNS caching disabled):
- MySQL: 400qps first run, 1030qps further runs
- PostgreSQL: 290qps for all runs
The following Postgresql settings where changed:
effective_cache_size
shared_buffers
The following Mysql settings where changed:
key_buffer
query_cache_*
Could you please let us know the exact values you've used?
Thanks
Norbert
--
Adam Cassar
IT Manager
NetRegistry Pty Ltd
______________________________________________
http://www.netregistry.com.au
Tel: 02 9699 6099 Fax: 02 9699 6088
PO Box 270 Broadway NSW 2007
Domains |Business Email|Web Hosting|E-Commerce
Trusted by 10,000s of businesses since 1997
______________________________________________
_______________________________________________
Pdns-users mailing list
Pdns-users@mailman.powerdns.com
http://mailman.powerdns.com/mailman/listinfo/pdns-users
