Which MySQL JOIN query is more efficient?

Question

Given the following table structure:

CREATE TABLE user (
   uid INT(11) auto_increment,
   name VARCHAR(200),
   PRIMARY KEY(uid)
);
CREATE TABLE user_profile(
          


        

Answer 1

The first syntax is generally more efficient.

MySQL buffers the derived queries so using the derived query robs the user_profile of possibility to be a driven table in the join.

Even if the user_profile is leading, the subquery results should be buffered first which implies a memory and performance impact.

A LIMIT applied to the queries will make the first query much faster which is not true for the second one.

Here are the sample plans. There is an index on (val, nid) in the table t_source:

First query:

EXPLAIN
SELECT  *
FROM    t_source s1
JOIN    t_source s2
ON      s2.nid = s1.id
WHERE   s2.val = 1

1, 'SIMPLE', 's1', 'ALL', 'PRIMARY', '', '', '', 1000000, ''
1, 'SIMPLE', 's2', 'ref', 'ix_source_val,ix_source_val_nid,ix_source_vald_nid', 'ix_source_val_nid', '8', 'const,test.s1.id', 1, 'Using where'

Second query:

EXPLAIN
SELECT  *
FROM    t_source s1
JOIN    (
        SELECT  nid
        FROM    t_source s2
        WHERE   val = 1
        ) q
ON      q.nid = s1.id

1, 'PRIMARY', '', 'ALL', '', '', '', '', 100000, ''
1, 'PRIMARY', 's1', 'ref', 'PRIMARY', 'PRIMARY', '4', 'q.nid', 10000, 'Using where'
2, 'DERIVED', 's2', 'ref', 'ix_source_val,ix_source_val_nid,ix_source_vald_nid', 'ix_source_vald_nid', '4', '', 91324, 'Using index'

As you can see, only a part of the index is used in the second case, and q is forced to be leading.

Update:

Derived queries (which is what this question concerns) are not to be confused with the subqueries.

While MySQL is not able to optimize derived queries (those used in the FROM clause), the subqueries (those used with IN or EXISTS) are treated much better.

See these articles in my blog for more detail:

• Passing parameters in MySQL: IN list vs. temporary table
• NOT IN vs. NOT EXISTS vs. LEFT JOIN / IS NULL: MySQL