[
https://issues.apache.org/jira/browse/JCR-314?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Marcel Reutegger updated JCR-314:
---------------------------------
Attachment: FineGrainedISMLocking.patch
ISMLocking.patch
To drive this issue forward I created a patch that replaces the current rather
hard coded use of a ReadWriteLock in SharedItemStateManager with an abstraction
layer that allows alternative implementations.
The way locking currently works is implemented in DefaultISMLocking and an
initial implementation of a more fine grained locking is available in
FineGrainedISMLocking.
FineGrainedISMLocking does not allow concurrent writes but at least allows
reads while a long running write takes place and the read does not conflict
with the write. I think this is the most common use case (save a big file but
still allow other sessions to read, which is currently not possible).
All tests pass with both the DefaultISMLocking and the FineGrainedISMLocking.
The first patch only includes the structural changes moving away from the hard
coded ReadWriteLock in the SharedItemStateManager to the ISMLocking interface.
It does not change the locking semantics that are currently in place. The
second patch is the proposal for a more fine grained locking strategy.
Comments are welcome.
> Fine grained locking in SharedItemStateManager
> ----------------------------------------------
>
> Key: JCR-314
> URL: https://issues.apache.org/jira/browse/JCR-314
> Project: Jackrabbit
> Issue Type: Improvement
> Components: core
> Affects Versions: 0.9, 1.0, 1.0.1, 1.1, 1.1.1, 1.2.1, 1.2.2, 1.2.3
> Reporter: Marcel Reutegger
> Attachments: FineGrainedISMLocking.patch, ISMLocking.patch
>
>
> The SharedItemStateManager (SISM) currently uses a simple read-write lock to
> ensure data consistency. Store operations to the PersistenceManager (PM) are
> effectively serialized.
> We should think about more sophisticated locking to allow concurrent writes
> on the PM.
> One possible approach:
> If a transaction is currently storing data in a PM a second transaction may
> check if the set of changes does not intersect with the first transaction. If
> that is the case it can safely store its data in the PM.
> This fine grained locking must also be respected when reading from the SISM.
> A read request for an item that is currently being stored must be blocked
> until the store is finished.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
