This set of datatypes represents an approximate value. The data stored in these types of columns is approximate because it depends upon the specific processor architecture and also the number of significant digits used in a calculation. The real datatype can store a wide range of values. Depending upon the specific processor architecture SQL Server is running on, you could find that you are not able to utilize the entire range of possible values.

Date/Time Data

SQL Server does not store dates and time separately. Both the date and time portions are stored in the same column. The first half of the storage space is used for the date and the second half for the time portion. Time is stored to an accuracy of 1/300^th of a second.

Since year 2000 issues always come up when discussing date data, it should be noted that SQL Server 7.0 is fully year 2000 compliant. The year 2000 problem mainly occurs in applications. Even though SQL Server is year 2000 compliant, if you couple it with an application that is not year 2000 compliant, you will have the same problems. If your application only passes 2 digit years, SQL Server will use a standard algorithm to determine which century to store the data as. If the value is between 0 and 49, then it will be stored as 2000 to 2049. If the value is between 50 and 99, then it will be stored as 1950 to 1999.

Money Datatypes

There are two specialized datatypes for storing monetary data. In most cases, a smallmoney datatype is sufficient. If you need to store values similar ot the US national debt, you will need to use a money datatype. You will note that these columns allow you to store values to 1/100 of a cent. Any data stored to this precision will calculate correctly, but you must be aware that many applications will round this off to two decimal places.

When storing information in these columns, you do not need to specify a monetary symbol. However, this does have an effect on the storage of the data. If you insert the value 6,134.73, it will be stored as a floating point number. If instead you insert the value $6,134.73 it will be stored as exact numeric value.

SQL Server 7.0 Datatypes 1 2 3 4 5 6