I've been using Hazelcast recently and because I quite like it, I'm at the risk of succumbing to The Law of the Instrument , or "to a hammer all problems are nails"
I'm wondering about how ridiculous it would be to store static final data, such as the list of countries for a web app because it is a kind convenient/lazy way to make it happen.
Given that HazelCast's design goal is really for distributed mutative state how appalling an idea would this really be and why?
If your static data doesn't fit in a single JVM memory then you can consider storing in Hazelcast but for small amount of data it will be over-engineering. Keep things local if you can to avoid the cost of possible serialization, networking, clustering etc.
founder@hazelcast