[jboss-user] [JBossCache] - Cache performance with millions of nodes

[jivesociety]

I would like some advice as to the best way to setup JBossCache with a FileCacheLoader in a clustered topology with synchronous replication. I would like to improve concurrency contention as much as possible. So, i was planning on having a key/value pair entry basically be its own node. The issue is that my project plans to have around 10 million entries in the cache eventually. We will be deploying to a Solaris environment and I've been running some tests putting large amounts of nodes in the file system and the read times are getting pretty bad when there are this many files in a directory.

Any advice on the best way to set this up?

thanks,
ray

View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=4011964#4011964

Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=4011964