On Sun, Nov 30, 2008 at 11:28 PM, Tom Lane <tgl@xxxxxxxxxxxxx> wrote:
"David Rowley" <dgrowley@xxxxxxxxx> writes:Indeed, work_mem is probably the problem. The critical difference
> I assume workmem, effective_cache_size and random_page_cost are all the same
> in the 2 postgresql.conf?
between the two plans seems to be that the first one is using a
"hashed subplan" and the second one isn't. Assuming the same datatypes in both databases, the only reason not to use a hashed subplan is if the hashtable is estimated not to fit in work_mem.
I changed work_mem in test machine to be implemented in production server later. Completely forgot that when faced with this issue. Sorry for the trouble.
I learnt about "hashed subplan" requiring sufficient work_mem, however.
Thanks and regards,
Ma Sivakumar
மா சிவகுமார்
எல்லோரும் எல்லாமும் பெற வேண்டும்
http://masivakumar.blogspot.com
