Hi,
I wasn't whether or not to mail to the novice mailing list of this one. Since this is performance related I'm posting it here, but I am definitely a novice at postgresql - converting from mssql just now.
I have a ~2.5gb table with ~5M rows of data. A query that groups by two fields and sums a floating field takes approximately 122 seconds. The equivalent query takes ~ 8seconds in my previous sql server express installation.
I've tried to vary the parameters in postgresql.conf:
I've tried wavering shared buffers from 512mb to 4000mb
and working_mem from 64mb to 4000mb (i thought this might be the answer since the execution plan (referenced below) indicates that the sort relies on an External Merge Disk method)
I've increased the default_statistics_target to 10000 and full vacuum analyzed
I realize there are no indexes on this table. My main concern is why I can't get this to run as fast as in sql server express (which also has no indexes, and the same query takes about 8 seconds)
My system: Windows Professional 64-bit
8 gb of ram
Intel i5-220M CPU @ 2.5GHz
Here is the link to the execution plan: http://explain.depesz.com/s/Ytx3
Thanks a lot in advance and do let me know if you require any more information to make an informed opinion,
A
