193

What is the MySQL command to retrieve the count of records in a table?

12 Answers 12

267
SELECT COUNT(*) FROM fooTable;

will count the number of rows in the table.

See the reference manual.

Sign up to request clarification or add additional context in comments.

9 Comments

Is it any faster when I use name of indexed column instead of * ? Like this: SELECT COUNT(id) FROM tablename
Only slightly. I got a 21961904 row table which his COUNT queried in 115 sec, while using the ID took 107 sec.
COUNT(*) is a strict language definition. You will get a parse error if you try COUNT (*) note the space
You can do COUNT(1), this will be the fastest way.
What is the best way for using COUNT() to write a variable in PHP? Do I do "...COUNT(*) AS rowCount..." in the SQL, does it use $results->num_rows, or is there a way to call this result directly?
|
104

Because nobody mentioned it:

show table status;

lists all tables along with some additional information, including estimated rows for each table. This is what phpMyAdmin is using for its database page.

This information is available in MySQL 4, probably in MySQL 3.23 too - long time prior information schema database.

The number shown is estimated for InnoDB and TokuDB but it is absolutely correct for MyISAM and Aria (Maria) storage engines.

Per the documentation:

The number of rows. Some storage engines, such as MyISAM, store the exact count. For other storage engines, such as InnoDB, this value is an approximation, and may vary from the actual value by as much as 40% to 50%. In such cases, use SELECT COUNT(*) to obtain an accurate count.

This also is fastest way to see the row count on MySQL, because query like:

select count(*) from table;

Doing full table scan what could be very expensive operation that might take hours on large high load server. It also increase disk I/O.

The same operation might block the table for inserts and updates - this happen only on exotic storage engines.

InnoDB and TokuDB are OK with table lock, but need full table scan.

4 Comments

This looks like the most efficient way to count rows. I wonder why it is not the answer? I'm using InnoDB. With table lock, the number of rows should be exact right?
for innodb it is estimated. but you can use it in some cases "Our website have xxx members", "We detected xxx results similar to yours" and so on.
I am not sure. but for Innodb is not real count. But is quite useful for million row tables.
Going through tables to count each row to get a row count is rather ridiculous. So I prefer this answer as well.
39

We have another way to find out the number of rows in a table without running a select query on that table.

Every MySQL instance has information_schema database. If you run the following query, it will give complete details about the table including the approximate number of rows in that table.

select * from information_schema.TABLES where table_name = 'table_name'\G

6 Comments

Is this faster as well on a MyISAM table since MyISAM is storing the number of rows?
I never tested this on MyISAM.
Why does someone want to see APPROXIMATE number of rows? WTF
@Raymond It's reasonable that someone may just trying to get an estimate for the sake of decision making, but doesn't need an exact count. In my experience, this is a more common scenario in a production environment, than an exact number.
@raymond when you work with big tables the exact number of rows is not a concern.
|
6

Simply:

SELECT COUNT(*) FROM `tablename`

Comments

5
select count(*) from YourTable

Comments

4

If you have several fields in your table and your table is huge, it's better DO NOT USE * because of it load all fields to memory and using the following will have better performance

SELECT COUNT(1) FROM fooTable;

2 Comments

This is incorrect. See stackoverflow.com/questions/5179969/… COUNT(1) may actually be much slower for MyISAM.
@jcoffland you can try t by myself, especially when you have a join or where conditions is will be much much faster than count(*).
4

If you have a primary key or a unique key/index, the faster method possible (Tested with 4 millions row tables)

SHOW INDEXES FROM "database.tablename" WHERE Key_Name=\"PRIMARY\"

and then get cardinality field (it is close to instant)

Times went from 0.4s to 0.0001ms

1 Comment

This is brilliant! Counted rows for an unusable table.
3

Just do a

SELECT COUNT(*) FROM table;

You can specify conditions with a Where after that

SELECT COUNT(*) FROM table WHERE eye_color='brown';

Comments

2

As mentioned by Santosh, I think this query is suitably fast, while not querying all the table.

To return integer result of number of data records, for a specific tablename in a particular database:

select TABLE_ROWS from information_schema.TABLES where TABLE_SCHEMA = 'database' 
AND table_name='tablename';

2 Comments

This may return an estimated number of rows. In one example, I got 55,940,343 rows using COUNT(*) but 56,163,339 using this method so it was off by more than 200k.
@jcoffland, I mentioned the answer by _ Santosh_ above which states the result is approximate. My answer is a more detailed practical query. Its quite clear across answers if you want precise count use count(*) with performance awareness
1
$sql="SELECT count(*) as toplam FROM wp_postmeta WHERE meta_key='ICERIK' AND post_id=".$id;
$total = 0;
$sqls = mysql_query($sql,$conn);
if ( $sqls ) {
    $total = mysql_result($sqls, 0);
};
echo "Total:".$total;`

1 Comment

This answer is outdated, mysql_result is deprecated in PHP 5.5.0 and removed in PHP 7.0.0
1

You have to use count() returns the number of rows that matches a specified criteria

select count(*) from table_name;

Comments

0

It can be convenient to select count with filter by indexed field. Try this

EXPLAIN SELECT * FROM table_name WHERE key < anything; 

Comments

Start asking to get answers

Find the answer to your question by asking.

Ask question

Explore related questions

See similar questions with these tags.