AdityaAgrawal007/hydra — reverse-engineered prompt

Build me a Java project called Hydra that works as a small distributed key value store, more like a learning version of Dynamo or Cassandra than a production database.

I want to be able to run a few nodes on my laptop, have them discover or know about each other, spread keys across nodes with consistent hashing, and let me put, get, and delete simple string values. Data should be stored on disk with a basic log structured storage approach, so if I stop and restart a node it can recover its data. Add a simple gossip style membership flow so nodes can notice other nodes joining or leaving.

Please include a clear README with how to build and run it, example commands for starting multiple nodes, and a tiny demo showing writes and reads. Keep the code understandable, with comments around the distributed systems parts. Add basic tests for routing, storage, and node behavior. Look up current docs online if you need to.