Actually, there will be creation of 2 new tables for each insert on
'primary' table, so for 10K records, we would have 20K tables. Those
tables each will never grow more than a few records each.
Is it better to have 1 monolithic table and have to search it, or small
individual tables but many of them?
Ron Johnson wrote:
>-----BEGIN PGP SIGNED MESSAGE-----
>Hash: SHA1
>
>On 03/15/07 22:14, louis gonzales wrote:
>
>
>>Hello List,
>>I want to write a statement-level trigger - one that happens once per
>>statement - such that, immediately after an insert into a table(which
>>gets a unique integer value as an ID from a defined sequence, being the
>>primary key on the table), a new table is created with foreign key
>>
>>
> ^^^^^^^^^^^^^^^^^^^^
>
>
>>constraint on that unique ID.
>>
>>
>
>So if you insert 10,000 records into T, you then have 10,000 new tables?
>
>-----BEGIN PGP SIGNATURE-----
>Version: GnuPG v1.4.6 (GNU/Linux)
>
>iD8DBQFF+nDLS9HxQb37XmcRAkrOAJ4pKzY/TVlm3K0aqiEY7ukhCfNZjwCfa5C7
>mukqcoAmU2/OYr4QFVxjt6k=
>=sHA3
>-----END PGP SIGNATURE-----
>
>---------------------------(end of broadcast)---------------------------
>TIP 6: explain analyze is your friend
>
>
--
Email: louis.gonzales@linuxlouis.net
WebSite: http://www.linuxlouis.net
"Open the pod bay doors HAL!" -2001: A Space Odyssey
"Good morning starshine, the Earth says hello." -Willy Wonka
