Re: Optimal time series sampling.

--- Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> wrote:

> Ted Byers <r(dot)ted(dot)byers(at)rogers(dot)com> writes:
> > OK, it is challenging to present it in plain text,
> but
> > here is the HTML exported by MySQL Query Browser.
>
> Why are you asking this list for help with a MySQL
> performance problem?
>
because my question isn't really about MySQL, but
rather about how best to construct the SQL required to
get the job done, regardless of what database is used.
I have seen some claims that it is better to use
joins instead of correlated subqueries and others that
say the opposite. And I do not, at this stage, know
if there are other options in SQL that may or may not
be better. At this time, the database in use is
irrelevant (I want to stick as close to the ANSI
standard as practicable so the rewriting required will
be minimal should we decide to change the database
later, for whatever reason).

Thanks,

Ted