Ron Johnson <ron.l.johnson@cox.net> writes:
> What do the indexes, if any, on tblData look like?
PRIMARY KEY (`id`),
KEY `idusername` (`id`,`username`),
KEY `zip` (`zip`(5)),
KEY `apptypezip` (`zip`(5),`apptype`(7)),
KEY `username_idx` (`username`(10))
on zipDistances
KEY `origdist` (`origin`,`distance`),
KEY `zip` (`zip_distant`)
> Also, remember that DISTINCT adds computational and disk
> overhead to the operation, since it has to *sort* all the
> records in the table, in order to eliminate duplicates.
Yes, but it is still /far/ faster than the really huge "zip in" that
we were using before.
And, it shouldn't lock the table, right?
> <ADVOCACY mode=FLAMEBAIT>
> This is why you should use a real RDBMS like PostgresSQL,
> rather than a hacked-up ISAM like MySQL.
> </ADVOCACY>
I totally agree. I plan to move this app to Postgres soon, but right
now I have to deal with this.
It gets worse. Instead of storing social security numbers in a
uniform format, queries check to see if the SSN is stored in one of a
number of formats. The same with dates (which are stored as varchar
fields).
So, yes, there is poor design here, but I still don't think that this
should cause the problem I'm seeing.
Mark.
-- A choice between one man and a shovel, or a dozen men with teaspoons is clear to me, and I'm sure it is clear to you also. -- Zimran Ahmed <http://www.winterspeak.com/> ___________________ Nolug mailing list nolug@nolug.orgReceived on 05/29/04
This archive was generated by hypermail 2.2.0 : 12/19/08 EST