"Luca Ferrari" <fluca1978@infinito.it> writes:
> Hi,
> sorry for this question, but as far as I know postgresql does not use a
> process pool, rather a new process is created for any connection on demand.
> If this is true, what is the reason for this?
Generally to add code the question is not why not, but why? That is, the
question you should be asking is why would we do it? What advantage do you
anticipate it would have.
I assume you think it would be faster. But forking processes on Unix is very
fast compared to starting a database session and transaction so the gain would
be minimal. And in any case you can always add a pool outside of Postgres
using things like pgpool or pgbouncer.
--
Gregory Stark
EnterpriseDB http://www.enterprisedb.com
Ask me about EnterpriseDB's RemoteDBA services!