Question

I have a mysql database with a table with about 40k entries. Executing the following statement takes about 10 seconds (Database is already selected):

SELECT * FROM MyTable WHERE Column < 3

Why does this take so long and how do I improve the performance?
Are other Databases faster? (e.g. MongoDB, CouchDB, … ) I’d prefer to use a MySQL Database though.

EDIT:

The following query…

EXPLAIN SELECT * FROM MyTable WHERE Column < 3;

results in the following:

id  select_type table   type    possible_keys   key key_len ref rows    Extra
1   SIMPLE  Occurances  index   NULL    SearchString    102 NULL    40242   Using where; Using index

The following query shows the distribution of values in the Column

SELECT COUNT(*), Column FROM MyTable GROUP BY Column;

The result is the following:

COUNT(*)    Column
43      0
5       1
106     2
71      3
42      4
283     5
2337    6
9491    7
22073   8
1191    9
1064    10
1105    11
919     12
393     13
288     14
288     15
200     16
123     17
71      18
71      19
36      20
10      21
13      22
8       23
4       24
3       25
4       29

Answer 1

I’d venture to guess that you have no index on the Column column. Try creating one:

CREATE INDEX idx_MyTable_Column ON MyTable (Column);

Try comparing the output of EXPLAIN before and after you create the index:

EXPLAIN SELECT * FROM MyTable WHERE Column < 3;

You should see that with the index, an index scan (or better) is performed.

An index will only help you if a small enough set of rows match your criteria. If most of the table matches the expression Column < 3 then an index won’t help and the planner will fall back on a table scan, since that will turn out to be faster than using the index.

---

If you want a more detailed answer, then you’ll have to provide more information. The output of these two queries would be helpful:

EXPLAIN SELECT * FROM MyTable WHERE Column < 3;
SELECT COUNT(*), Column FROM MyTable GROUP BY Column;

As well as a list of indexes on MyTable.