I am currently using EHCache in my application and I am investigating a potential move to JBossCache. The main reason is that we need some kind of cache replication.
EHCache provides the following feature:
* You can define the maximum number of element stored in the memory store.
* If the memory store reaches this maximum number of elements, the objects are stored in the 2nd level cache, which is basically a disk store.
This is great since it allows us to control the amount of memory used by the cache, BUT on top of that, the elements that cannot be stored in memory are still available via the disk store.
Is there something similar in JBoss Cache ?
I had a look at the eviction policies. As far as I understand it, the maxNodes setting applies to the both the nodes stored in the memory cache and to the nodes stored in the cache loader. Therefore, this means that you cannot set a maximum number of nodes stored in the in-memory cache.
Is my understanding correct ?
Thanks a lot,
This is currently being designed/implemented: http://jira.jboss.com/jira/browse/JBCACHE-66. There is already a design document, and Hany will start a public discussion on the JBossCache forum soon.
This should be available in the 1.2.5 release (Oct 2005)