Good question. I once dug through this a little in BOL to understand all of the data types and the differences. Since then, I use bit, small int, varchar always unless it is a Social Security Number or 1 character, small datetime, etc. It is important to understand how the data is stored both for storage space reasons as well as index performance. For example. Lets say you have a column in a table that is used for display order of the records, and you know it will never be more than 10. Look at below what I pasted in from BOL. Tinyint is the way to go in this case.
bigint Integer (whole number) data from -2^63 (-9223372036854775808) through 2^63-1 (9223372036854775807). int Integer (whole number) data from -2^31 (-2,147,483,648) through 2^31 - 1 (2,147,483,647). smallint Integer data from 2^15 (-32,768) through 2^15 - 1 (32,767). tinyint Integer data from 0 through 255. Dan -----Original Message----- From: Michael Hodgdon [mailto:[EMAIL PROTECTED] Sent: None To: CF-Talk Subject: WOT: SQL Data types I am curious how religiously people stay true to using data types for MS SQL Server. I know some of these stretch to other RDBMS's, however, I would like to know what people in industry think. What type of general rules of thumb do people follow when planning data types for a data store? For instance Money, Numeric, varchar, int, bigint ... I think you get the point. I am looking for response such as; never use that type, it won't convert to Oracle. Never use that type, the space you save is not worth it, etc.... If this needs more elaboration, please let me know Mike. ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~| Archives: http://www.houseoffusion.com/lists.cfm?link=t:4 Subscription: http://www.houseoffusion.com/lists.cfm?link=s:4 Unsubscribe: http://www.houseoffusion.com/cf_lists/unsubscribe.cfm?user=89.70.4 Get the mailserver that powers this list at http://www.coolfusion.com
