The floating point data type consists of numeric data with decimal points and/or exponents.

Floating point data can contain an integer portion, a decimal point, and a fractional portion and/or scientific notation, in this format:

For example:

You can change the default decimal point indicator, the period (.), to another character, such as the comma (,), used in European decimal format, by using the II_DECIMAL system variable. For more information, see the System Administrator Guide and the set decimal command in your query language reference guide.

On output, a mantissa of 1 is supplied when an exponent is missing a mantissa.

The floating point range and precision vary greatly among the various operating systems and hardware.