Chris Fossenier writes:
... Query 1
> a1.phone_pander_flag <> 'Y'
> AND state.state = 'PA'
> AND ( h1.homeowner = 'Y'
> OR h2.probable_homeowner IN ('8','9')
> OR h2.homeowner_probability_model BETWEEN '080' AND '102' )
> AND ( p1c.exact_age BETWEEN '40' AND '60'
> OR estimated_age BETWEEN '40' AND '60' )
> AND a1.phone is not null
> AND p1a.first is not null
> AND p1a.last is not null
> AND a3.address is not null;
Chris,
Have you been able to get an 'explain plan' on the query? I'd
estimate that the combination of state (1 of 50), exact_age (20 of 80),
and estimated_age (20 of 80) should reduce you possible record search
from the original 160M to 200,000 (1/50 * 20/80 * 20/80 * 160M) if all
these indexes can be utilized. The real question is what MySQL chooses
to use as far as indexes are concerned.
Brad Eacker ([EMAIL PROTECTED])
--
MySQL General Mailing List
For list archives: http://lists.mysql.com/mysql
To unsubscribe: http://lists.mysql.com/[EMAIL PROTECTED]
