Cassandra provides a write consistency, which is unique and almost never used in any situation. This consistency level is ANY. In this case, the write request is successful even if all the replica nodes are down for that partition key. To check this consistency level in action, let's bring down another node.
Before bringing down a second node, change the consistency to ONE, and execute the third query from the preceding section. You will see that it executed successfully, and a SELECT query will return the latest results:
Now that you set the consistency level to ONE, every request requires only one replica node to acknowledge ...
