Optimizing this query....

Question

I have a query for the following
select x.date_required, sum(x.t_score) as t-score from
(select tv.business_date as date_required, sum(tv.score) as t_score from table1 tv
where tv.type like 'blue'
union all
select ta.business_date as date_required, sum(ta.score) as t_score from table1 ta
where ta.type like 'White'
union all
select tg.business_date as date_required, sum(tg.score) as t_score from table1 tg
where tg.type like 'green') x
group by x.date_required;
I would like to optimize it to make it simplier. Thank you.

Answer 1

Hi,
There are errors in the query you posted, so it's hard to know what you're trying to do.
For starters, find a way to avoid the UNION. For example:
SELECT        business_date          AS date_required
,       SUM (score)          AS t_score
FROM       table1
WHERE       type     IN ('blue', 'white', 'green')
GROUP BY  business_date
;If you're not using wild-cards, don't use LIKE; it's slower than = or IN.
Do you have an index on type?
The following thread show what you need to do for optimization questions:
When your query takes too long ...