This is an open end question, the interviewer says this is to test my thinking process.\n\nSuppose it has BST of data (does not matter what it is) in a node that is being updated concurrently, you need to copy this BST to another node.  How do you do it?\n\nI really have no experience in data replication.   I say to update the BST concurrently, it must have to lock it.   So, to copy the BST to another node, it must have a way to do distributed lock.  We then start to discuss how to do a distributed lock.  He then says suppose there is a master node that will never fail and all other nodes can go up or down, how can I implement distributed lock?   I say it can have hashtable in a NoSQL to tell if a node is locked by which one, but he then ask when happen if the node is down and how to unlock it.   I really don\'t know what I am talking about, just say to say something.\n\nHe is the hiring manager and I got rejected next day.  :(\n\nI really have no clue how to answer this question.   Any suggestions or comments are welcome.

This is an open end question, the interviewer says this is to test my thinking process.\n\nSuppose it has BST of data (does not matter what it is) in a node tha