You use the data type FLOAT for columns that can store floating-point numbers (see section "Numeric values"). The precision can be specified.
[{[ dimension ] | ( dimension )}] FLOAT[( precision )]
dimension
Unsigned integer between 1 and 255. The column is a multiple column;
dimension indicates the number of column elements.
dimension omitted:
The column is an atomic column.
precision
Unsigned integer between 1 and 53 that indicates the minimum number of binary digits for the mantissa.
precision omitted:
precision=1.
Range of values for FLOAT columns
A FLOAT column can contain a floating-point number whose value is 0 or ranges from
5.4E -79 to 7.2E +75.
In SESAM/SQL, the precision of FLOAT floating-point numbers is 53 binary digits if precision is greater than 21, otherwise it is 21 binary digits.
Example
You define a FLOAT column test_value with a precision of at least 30 binary digits as follows:
test_value FLOAT(30)