Numeric Data Types

Complete numeric types reference: INT/BIGINT ranges, DECIMAL(M,D) precision/scale, FLOAT/DOUBLE storage sizes, and numeric type selection guidelines.

Numeric Data Type Overview

General introduction to numeric data types. This page summarizes the available integer, fixed-point, and floating-point types and their storage characteristics.

BIGINT

Large integer type. A BIGINT uses 8 bytes and can store values from -9223372036854775808 to 9223372036854775807 (signed) or 0 to 18446744073709551615 (unsigned).

BIT

Bit-field data type. A BIT(M) column stores M bits per value, allowing storage of binary values from 1 to 64 bits in length.

BOOL

Synonym for TINYINT(1). This type is commonly used to represent boolean values, where 0 is considered false and non-zero values are true.

BOOLEAN

Complete BOOLEAN type reference: TINYINT(1) synonym, TRUE/FALSE aliases, CREATE TABLE/SHOW CREATE TABLE output, and IF()/IS operator evaluation.

DEC, NUMERIC, FIXED

Synonyms for DECIMAL. These keywords declare fixed-point numbers, which store exact numeric data with a defined precision and scale.

DEC

Synonym for DECIMAL. This keyword creates a fixed-point column with exact precision, suitable for financial calculations.

DECIMAL

Complete DECIMAL type reference: DECIMAL(M,D) syntax, precision limits (M≤65, D≤38), SIGNED/UNSIGNED/ZEROFILL options, and rounding behavior.

DOUBLE PRECISION

Synonym for DOUBLE. This keyword declares a normal-size (8-byte) floating-point number with double precision.

DOUBLE

Double precision floating-point number. A DOUBLE column uses 8 bytes to store large or precise approximate values.

FIXED

Synonym for DECIMAL. This keyword is used to define columns that require exact numeric precision, such as currency.

FLOAT

Single precision floating-point number. A FLOAT column uses 4 bytes and stores approximate values with less precision than DOUBLE.

FLOAT4

Synonym for FLOAT. This keyword declares a single-precision floating-point column.

FLOAT8

Synonym for DOUBLE. This keyword declares a double-precision floating-point column.

Floating-point Accuracy

Explanation of floating-point precision issues. This page details why FLOAT and DOUBLE types are approximate and how rounding errors occur.

INT

Complete INT type reference: INT(M) syntax, signed (-2147483648 to 2147483647) vs unsigned (0 to 4294967295) ranges, and ZEROFILL padding guidelines.

INT1

Synonym for TINYINT. This type uses 1 byte of storage.

INT2

Synonym for SMALLINT. This type uses 2 bytes of storage.

INT3

Synonym for MEDIUMINT. This type uses 3 bytes of storage.

INT4

Synonym for INT. This type uses 4 bytes of storage.

INT8

Synonym for BIGINT. This type uses 8 bytes of storage.

INTEGER

Synonym for INT. This keyword declares a standard 4-byte integer column.

MEDIUMINT

Medium-sized integer. A MEDIUMINT column uses 3 bytes and stores values from -8388608 to 8388607 (signed) or 0 to 16777215 (unsigned).

MIDDLEINT

Synonym for MEDIUMINT. This keyword refers to the 3-byte integer type.

NUMBER

Oracle-compatible synonym for DECIMAL. This type is used for fixed-point arithmetic.

NUMERIC

Synonym for DECIMAL. This type stores exact numeric data values.

REAL

Synonym for DOUBLE. In standard SQL mode, REAL is a double-precision floating-point number.

SMALLINT

Small integer type. A SMALLINT column uses 2 bytes and stores values from -32768 to 32767 (signed) or 0 to 65535 (unsigned).

TINYINT

Very small integer type. A TINYINT column uses 1 byte and stores values from -128 to 127 (signed) or 0 to 255 (unsigned).

_{This page is licensed: CC BY-SA / Gnu FDL}