Finding a XML-Database to fit our needs

[Ken North]

Michael Kay wrote:

>> The database literature is full of alternative techniques.
>>... three techniques were recognized: record-level locking (which I guess
>>equates to node-level in an XML database), page-level locking, and predicate
>>locks. Fine-granularity locking looks good on paper but the benefits can
>>often be lost because you spend too much time doing lock acquisition/release
>>and deadlock detection ...

Which is why yet another alternative was developed. Multi-versioning was 
InterBase's claim to fame and a key reason Borland acquired it for the 
client-server database wars of the '90s. (InterBase is now Firebird and 
Microsoft SQL Server now provides multi-versioning.)

Clearly we need something more granular than document-level locking, 
particularly for very large documents.

The obvious question -- is the capacity to handle terabyte-sized documents 
essential for this application?
Is there no basis for partitioning the data to permit better concurrency and 
parallel execution of queries? Oracle XML DB, for example, supports parallel 
query and a separate, transportable tablespace for XML repositories. Likewise 
DB2 and SQL Server have mature technology for parallel execution.