SQL Server 2005

[Ronald Bourret]

Frank Cohen wrote:

>  I'm
> working on a performance and scalability study that among other  things 
> compares performance of native XML DB tools to relational  tools that 
> will back up a critique.

I'm curious about the data set you'll use for this.

As a general rule, it seems that the use cases for XML-enabled databases 
and native XML databases are separate. In particular, the general rule 
of thumb is that, if the data easily maps to the relational model, then 
do that. If it doesn't, start thinking about using the XML data model 
directly.

And yes, I know that a lot of real-world cases mix relational and XML 
data, but even that still fits the above rule -- shred into the 
relational model until you hit something that doesn't fit (like a prose 
description of a part) and then store that as XML. This assumes that 
document-centric data is embedded in an otherwise data-centric document, 
but that situation seems far more common than the reverse.

-- Ron