On Sat, 2005-09-03 at 00:59 -0800, Poul Jensen wrote:
> I'm building a database containing key parameters for ~500,000 data
> files. The design I found logical is
>
> Two tables for each file:
> 1) Larger table with detailed key parameters
> (10-15 columns, ~1000 rows), call it large_table
> 2) Small table with file summary
> (~30 columns, 1 row), call it small_table
you want to create 1 million tables, all with one of
2 schemas?
why not just 2 tables, each with the additional "file"
column ?
> ...
> SELECT <large_table columns> FROM <regular expression>
> WHERE <condition on large_table>
> IF <condition on corresponding small_table>;
this would then be something like:
SELECT <large_table columns> FROM large_table
WHERE file ~ <regular expression>
AND <condition on large_table>
AND <subquery involving small_table>
gnari