Обсуждение: ODBC driver performance on large result-sets

Attached are two C programs. They borh (I believe) execute the same
query against a PostgreSQL database on my WinXP laptop. The libpq
version runs quickly (in a second or less), while the ODBC version
takes about 40 seconds. Also, while the ODBC version is processing,
the memory usage for one of the postgres server processes goes up
quite a bit, to 140M.

Any suggestions as to what's causing the difference? And, is there
something I can do about it?

Thanks,
Alistair

> Attached are two C programs. They borh (I believe) execute the same
> query against a PostgreSQL database on my WinXP laptop. The libpq
> version runs quickly (in a second or less), while the ODBC version
> takes about 40 seconds.

Never mind. I just noticed that the ODBC driver has "disable genetic
optimizer" checked. If I uncheck then presumably the query gets some
decent optimisation, and runs as quickly as the libpq version. Is this
a sensible default setting?

Sorry for the spam.

Alistair