XML.com: Lightweight XML Search Servers, Part 2 [Feb. 18, 2004]

XML.com: Lightweight XML Search Servers, Part 2 [Feb. 18, 2004]: "This month's challenge was to scale up to a database-backed implementation using Berkeley DB XML."

I don't know if I'd use BerkeleyDB XML for this, I'm unsure of how fast it's XPath queries are; I know that it's locking is document-level, and I'd prefer Node-level locking. (Is Node-level locking possible?)