Large XML documents ... RDB or Parse?

Forgive me if this sounds naive, but I want to find out if there is another option before jumping entirely in one direction.

I am working on a project using a Flash/PHP interface for a large XML document and maybe a web service. Not much problem with the basics, I've been doing the like for the past year. This time, however, the document in question is over 7 MB.

My first obvious thought is to rip the entire document into a relational database that reflects the structure of the document, then serve up just requested portions in the original format.

My question is, is this necessary or is there some way I can serve directly from the document without taxing server resources? It sounds to me like a pipe dream to do so, but here my nievite may shine as I've never worked with things on such a scale.

So, do I go RDB or can someone suggest the direction I should look in order to manage it from the source doc?

Chris Hemming
Digital Stone Productions
URL

hi,
try sax or stAX
thanks,
with regards,
subramanian.T