Here is a perplexing issue I am having:
Query:
EXPLAIN SELECT id,hostname FROM queue_servers WHERE live=1
id select_type table type possible_keys ke
Have you tried enforcing a particular index for search ? http://dev.mysql.com/doc/refman/5.0/en/index-hints.html
You can tell the DBMS to use a proper index for the query. That should give you predictable behaviour.
Why doesn't MySQL not use an index?
MySQL will not use an index if a large percentage of the rows have that value.
Why will adding use index
to the query not work here
Adding a use index
clause will have no effect, because use index
will only suggest which index to use, it will not suggest whether to use an index or not.
Caveat when using test tables with few rows
This is especially vexing when using test tables with few rows as MySQL will refuse to use an index, and it's hard to see what's wrong with your query.
So make sure you add enough rows to a test table to make it a realistic test.
Is using an index on low cardinality columns useless?
Indexing on boolean columns is not as useful as you thought before asking this question.
However it is also not useless either.
With InnoDB MySQL will try and retrieve data using the indexes if possible, if your boolean field has an index the query:
SELECT id, bool_field FROM table_with_many_columns_and_rows WHERE bool_field = 1
Can read all the data in the covering index for bool_field
because secondary indexes in InnoDB always include the primary index (id) as well.
This is faster because MySQL does not have to read the entire table into memory.
In MyISAM this doesn't work and MySQL will examine the whole table.
But I can use force index
You can, but on low cardinality indexes it will make your query slower, not faster.
Only override the indexes on complex queries and only if you know the rules MySQL uses to select indexes.
Links:
See: http://dev.mysql.com/doc/refman/5.1/en/mysql-indexes.html
and: http://www.xaprb.com/blog/2006/07/04/how-to-exploit-mysql-index-optimizations/
If you want a book on this subject: read
High performance MySQL: http://oreilly.com/catalog/9780596003067