How to optimize MySQL query which contains table and left join of sql view

Question

I have this query:

SELECT a.id, b.discount FROM (a LEFT JOIN b ON b.id_a = a.id);

EXPLAIN command for this query is this:

id  select_type   table  type   possible_keys   key       key_len  ref    rows   Extra
1    SIMPLE        a      index  NULL            PRIMARY   4        NULL   1489   Using index
1    SIMPLE        b      ALL    NULL            NULL      NULL     NULL   819   

A is table with 1489 rows and B is a SQL view. Now I understand, that MySQL has to do 1489*819 operations which is way too not-optimized operation. If B was a table, I would create an index for column id_a, but I have no idea what to do with SQL view.

Anybody can help ?

Answer 1

I would look at the view, and try to go to the raw tables that qualify the element you are concerned with finding exists or not. And ensure the root table of "b" view has an index on that key.

Answer 2

Instead of working with a view (which is unindexed), you could also try creating a temporary table with indexes. We did that for some lengthy reporting queries and got some impressive speed improvements as a result (YMMV of course).

Answer 3

Switching the table and view might help.

example:

SELECT a.id, b.discount FROM (b RIGHT JOIN a ON b.id_a = a.id);