Cache vs HashMap for simple usecase

This must be a very basic:- Just curious, If I don't need distributed, cache-as-sor models, why do we need third party cache libr开发者_开发问答aries (ehcache, memcached) when all you need (for simple use case) is just a key-value pair holder, something like HashMap ?

A lot of thought goes into producing software, and the more thought and testing by others (and fixes) improves the value of the software and also validates the code as a model (I didn't say a good model).

For the example, above, how would you handle the deleting of "old" cache items? You would have to add more code/features to insure that the cache could be emptied.

Using memcache may be overkill for a simple program, but it's already solved many of the problems that you will have and gives you a bit of extra ability.

I would also use Redis as an example. You can DO a lot of stuff in your own language, but sometimes, Redis would make other items easier.

YMMV!

-daniel