MySQL Count (*) and Count (column) rate

2. COUNT (*) is typically an index scan of a primary key, and Count (column) does not necessarily
4. COUNT (*) is the total number of records in the statistics table, count (column) is the number of records for all eligible columns in the tables
6. Count, if there is no where limit, MySQL directly return to save the total number of records, and there is a where limit, always to the MySQL full table traversal
8. If the column in Count (column) is a primary key, Count (column) is faster than COUNT (*), otherwise, COUNT (*) is fast
10. In any case, select COUNT (*) from table is the best choice
12. Minimize select COUNT (*) from table where column = ' value ' such a query
14. Eliminate the appearance of select Count (column) from table
16. The offset of the column determines the performance, the higher the column, the greater the cost of access, the count (*) algorithm and the offset, so count (*) the fastest, count (the last column) slowest

MySQL Count (*) and Count (column) rate