On Wed, Jul 25, 2012 at 1:40 AM, AI Rumman <rummandba(at)gmail(dot)com> wrote:
> Thanks. I missed to add the trigger.
> Now I added it, but still without partition taking less time compared to
> with partition query.
Based on the different times on "Seq Scan on table2", it looks like
one query has better caching than the other.
Did you try running the queries in alternating order, to average out
caching effects?
Could you run the "explain (analyze, buffers)" on those to get a
better picture of the buffer effects?
Cheers,
Jeff