Execution sequence of IN and JOIN in MySQL

Question

I have a SQL query which looks like the following.

SELECT A.a, count(B.id)
FROM TableA A inner join TableA B on A.referId = B.id
WHERE A.id in (123,2424,232...)
GROUP BY A.id

While executing this query

The rows are filtered using where clause first and then join is performed or
First join is performed, then the rows are filtered?

I am running this query on MySQL Server. Assuming TableA contains million rows. Is there any alternative and efficient way of writing this query?

Depends. The optimizer is supposed to pick the fastest way, depending on data amount, keys etc. — jarlh, Jun 03 '15 at 12:57
@RaviChandra I see that now :). Is there an index on `referId`? I assume `id` has an index. — Michael Berkowski, Jun 03 '15 at 12:58
we don't have index on referId. creating one will help gain performance? — Ravi Chandra, Jun 03 '15 at 12:59

score 1 · Accepted Answer · answered Jun 03 '15 at 13:14

The order of execution is determined by the query planner.

You can influence it by creating indexes.

For this query I suggest changing COUNT(b.ID) to COUNT(*). I also suggest creating a compound index on (id, referId). That should provide coverage for the stuff you need from the first instance of your table.

For more information on query performance, this is an excellent resource. http://planet.mysql.com/entry/?id=661727

score 0 · Answer 2 · answered Jun 03 '15 at 13:24

The rows selected by a query are filtered first by the FROM clause join conditions, then the WHERE clause search conditions, and then the HAVING clause search conditions. Inner joins can be specified in either the FROM or WHERE clause without affecting the final result.

Execution sequence of IN and JOIN in MySQL

2 Answers2