Hello,
We are looking for fast and simple Java XML database engine for our integration solution. I’ve found BaseX that seems to be a good option but I have some questions about it:
-
Performance. How it feels on large databases, about 50-100 millions of nodes? Does performance degrades significally? What is the maximum number of nodes that BaseX have been tested with?
-
Streamable queries. We will need to send large datasets to client including the whole database. Does BaseX provides a streamable queries without caching whole resultset nowhere in memory (something like server
cursors in RDBMS)? If we will use XQJ API and Java as middleware, is it will possible to create a fully streamable service to retrieve large amounts of data directly from database?
-
Failover. Our clients strongly require failover mode. I saw that at the moment BaseX does not support clustering and replication but it can be done at OS level using distributed file systems like Ceph or ClusterFS.
Can we have two o more BaseX instances connected to the same shared storage?
-
What transaction isolation levels BaseX supports? Do BaseX transactions support JTA architecture?
Regards,
Antón