Hi Pascal,
since Version 6.7.2 or 7.0, we have added a bulk mode to BaseX, which speeds up the insertions of new documents quite a lot [1]. What you have to do:
- set the "autoflush" option to false - perform all the insert operations - perform an explicit "flush" command
The test code listed at [2] might show how to properly use the flag.
Hope this helps, Christian
[1] http://docs.basex.org/wiki/Databases#Manage_Resources [2] https://github.com/BaseXdb/basex-tests/blob/master/src/test/java/org/basex/t... ___________________________
On Tue, Nov 15, 2011 at 4:38 PM, Pascal Heus pascal.heus@gmail.com wrote:
All: We often need to load a large number of small XML documents in a BaseX database. These are not particularly readily serialized in files (actually are in memory Apache XML Beans). Right now we're combining them in one "document" and perform a single insert call. We would though rather keep them as separate documents but this would entail calling insert thousand of times. Any suggestion as to how this could be performed efficiently (like fast/bulk inserts or through post processing)? Also, is there a significant different performing inserts using direct database access vs client/server (on localhost)? best Pascal
BaseX-Talk mailing list BaseX-Talk@mailman.uni-konstanz.de https://mailman.uni-konstanz.de/mailman/listinfo/basex-talk