>Yes, I believe that long transactions can't work without Castor's internal
>caching. Regarding JCS, are you using it for clustering? One of the
>features that is being planned for the next major version of Castor is a
>plugable (distributed) cache architecture that supports clustering. Would
>you be interested in sharing your insights about how we might be able to
>use JCS with Castor, and perhaps like to help out?
Our JCS use is pretty simple - we cache oql results by the Primary Key.
This does result in a lot of duplicate objects in the cache.
We prefer JCS cache since
- This was the big one - it allows us to flush the cache - we
need to flush different cache regions at different times of the day to
allow for other inputs to our data sources - we have plugged in a
timer mechanism to work with JCS.
- We are working on building JMX instrumentation to flush the cache
on demand.
- You have the option of caching to disk.
We use JCS as a local cache for each appserver instance though we will be using distributed caching at some point.
I would like to help out - I will ping you offline about that.
Thanks,
Mehul.
