As a general practice, we should use the most precise type for storing data. For example, a CHAR data type should be used to store a string value that varies in length from 1 to 255 characters. Another example is that MEDIUMINT UNSIGNED should be used to store numbers ranging from 1 to 99999.
Basic operations such as addition, subtraction, multiplication, and division with DECIMAL data are performed with the precision of 65 decimal digits.
Based on the importance of accuracy or speed, use of FLOAT or DOUBLE should be chosen. Fixed point values stored in BIGINT can be used for higher precision.
These are general guidelines, but the decision to use the right data type should be made based on the detailed characteristics explained separately for each data type in the earlier sections.
