Hi, >>For the current MongoMK implementation, it's the "sync" collection. > >Is that conceptually equivalent to the journal in Jackrabbit 2.x ?
In a way, yes. Actually there is also a (currently serialized) journal in the MongoMK, same as in Jackrabbit 2.x. I just noticed we didn't define scalable writes as a goal: http://wiki.apache.org/jackrabbit/Goals%20and%20non%20goals%20for%20Jackrab bit%203 - we have "High write throughput across cluster nodes" and "Concurrent writes within single cluster node". In my view, we simply forgot to explicitly add "Concurrent, scalable writes within multiple cluster nodes" as a goal. Unless somebody disagrees of course. Regards, Thomas
