Question

SELECT COUNT(*) FROM BigTable_1

Which way I should to use to get number of rows in table if I have more than 1 billion of rows?

UPDATE: For example, if we have ‘a timeout problem’ with the query above is there any way to optimize it? How to do it quicker?

Answer 1

If you need an exact count, you have to use COUNT (*)

If you are OK with a rough count, you can use a sum of rows in the partitions

SELECT SUM (Rows)
FROM sys.partitions
WHERE 1=1
And index_id IN (0, 1)
And OBJECT_ID = OBJECT_ID('Database.schema.Table');

If you want to be funny with your COUNT, you can do the following

select COUNT (1/0) from BigTable_1