Data Warehouse Logical Schema Name and Value Conventions
This topic describes the overall Data Warehouse and data value conventions.
Conventions for Class/Member/Key/Relation Names
Characters
The first character must be one of the following:
A letter as defined by the Unicode Standard version 2.0. The Unicode definition of letters includes Latin characters from a through z and from A through Z, and letter characters from other languages.
The underscore (_), at sign (@), or number sign (#).
Certain symbols at the start of an identifier have special meanings in SQL Server. An identifier that begins with the at sign indicates a local variable or parameter. An identifier that begins with a number sign indicates a temporary table or procedure. An identifier that begins with double number signs (##) indicates a global temporary object.
Some Transact-SQL functions have names that begin with double at signs (@@). To avoid confusion with these functions, we recommend that you do not use names that begin with @@.
The next characters can be:
Letters as defined in the Unicode Standard 2.0.
Decimal numbers from either Basic Latin or other national scripts.
The at sign, dollar sign ($), number sign, or underscore.
The identifier must not be a Transact-SQL reserved word. SQL Server reserves both the uppercase and lowercase versions of reserved words.
Size
Class names created through the provider in FastLoad mode are limited to 57 characters.
Member names created through the provider in FastLoad mode are limited to 100 characters.
Key names are limited to 128 characters.
Relation names are limited to 128 characters.
Conventions for Class/Member/Key/Relation Values
Key values should be less than 450 wide characters. SQL Server requires key values to be less than 900 bytes. This is equivalent to 450 wide characters.
A key can have up to 16 members.
Currency member values should use only the decimal point. Comma-separated values and locale symbols, such as $, are not allowed.