Archives of the TeradataForum

Message Posted: Thu, 22 Oct 2009 @ 13:34:22 GMT

	<Prev	Next>		<<First	<Prev	Next>	Last>>

I once consulted on a SQL Server DW (many issues found, design related), which had a table modeled with a multi-column key to allow duplicate rows. The multi-column key would be ok, , but the designer had inserted a "faux key" column, which supported some strange mainframe data to be double loaded, i.e. duplicate rows loaded, but with the odd data in the faux key column. It was strictly a workaround, but unnecessary and it created query and loading issues. We converted the table to 3NF, removed the dupes, and created another relational table to capture the odd mainframe data. Other than that, I can't really envision a real need for duplicate data rows.

	<Prev	Next>		<<First	<Prev	Next>	Last>>