From http://blogs.tedneward.com/2006/06/26/The+Vietnam+Of+Computer+Science.aspx
"developers usually use one of the other two approaches, more complex in perspective, but more efficient when working with relational storage: they either create a table for a particular (most derived) class, preferring to use denormalization and its costs, or else they create a single table for of the entire hierarchy, often in any case create a discriminator column to indicate which class each row in the table belongs to. (Other hybrids of these schemes are possible, but usually do not produce results that are substantially differ from these two.) Unfortunately, the denormalization costs are often significant for a large amount of data, and / or the table will contain significant amounts of empty columns, which will require NULLability restrictions for all columns, excluding powerful integrity constraints,proposed RDBMS. "
I would like to know which ORMs use one or the other way, especially those that use the same table and .NET?
source
share