Hi all, http://kohlerm.blogspot.com/ <http://kohlerm.blogspot.com/2008/05/analyzing-memory-consumption-of-eclipse.html>
thats me :) Actually I have a lot of experience with avoiding duplicated Strings and so far it almost always turned out to be a good thing (for performance and scalability) to avoid those duplicates. The implementation String.intern on modern JVM's is not that bad anymore, probably not as good as our (SAP) optimizations that we build into our own JVM. Those interned Strings should be long lived and therefore is not a big problem that they will live in permspace. It's pretty easy to build something that can work as replacement for String.intern, but i it's not that easy to do it without a big memory overhead. The problem is that we would need specialized implementation of a ConcurrentWeakHashSet for Strings. The biggest problem is that WeakReferences itself need quite a bit of memory. If you need help with this topic just let me know. I have a new command ready for the Memory Analyzer that will simplify finding those duplicates very much. Regards, Markus Regards, Markus
_______________________________________________ eclipse-incubator-e4-dev mailing list [email protected] https://dev.eclipse.org/mailman/listinfo/eclipse-incubator-e4-dev
