I'm looking at using TreeCache as a distributed cache in a project I'm currently working on.
Quesiton is; if I have two business tier app servers where I want to cluster TreeCache across the two, but also have two web servers where I want to use the cached objects placed there from business, but do not want those servers to store local copies of any cached data, how would I go abouth that? (Hope that is vaguely understandable)
Essentially; I need to be able to look up a cache instance on one(don't care which) business tier server form the web tier. I still want to replicate the cached objects between the two business tier servers for failover, but don't want a local cache on the web tier.
Can I do a JNDI lookup from web to business and keep a reference to the cache?
The reason that the web servers should not be a part of the cache cluster is because someone has decided they want them to be as 'light' as possible and so we need to explore alternatives to replicating the object to those servers.
This is to sit in a Webpshere environment.
Thanks for your time,