consistent_hashing

consistent hashing algorithm implementation in golang

overview

Basically consistent hashing can be used both as load balancer(It majorly treats here as stateless) and in databases side distribute the data and route each request to proper node. In case of databases we need to be more carefully like on addition of node we need to re-allocate data also to this node from other node and in same way of node eviction . Current implementation is only for load balancing.

Key Topics

1. We have used crc32 hashing library , but its extendable you could use any one just implement that logic.
2. AddNode is for adding new node
3. replicationfactor is number of copies of same node need to be added
4. Get is for getting nearest node to serve a request

Future Extension

Implement this logic for databases , which would be more challenging and more fun.