sdv mailer wrote:
> Instead, there's a big need to
> create a new connection on
> every query and with PostgreSQL needing to fork on
> every incoming connection
> can be quite slow.
Really? My general experience has beent that forking/connection setup
times are very good with PgSQL. Do not assume your Oracle experience
transfers directly over -- Oracle has very large connection time
overheads, PgSQL does not.
> This could be a big win since even a moderate
> improvement at the connection
> level will affect almost every user. Any chance of
> that happening for 7.5?
Only if you do it yourself, probably. The calculation of the developers
appears to be that the amount of time spent by the database on
fork/connect will generally be dwarfed by the amount of time spent by
the database actually doing work (this being a database, the actual
workloads required of the backend are much higher than, say, for a web
server). So the operational benefit of adding the complexity of a
pre-fork system is not very high. And if you have the rare workload
where a pre-fork actually *would* speed things up a great deal, you can
solve the problem yourself with a connection-pooling middleware.
-- __ / | Paul Ramsey | Refractions Research \_
