Whether we are reading or writing, the first thing that needs to happen is that we need to locate where the data should live on the system. This requires some type of key-to-node mapping.
In Dynamo, keys are mapped to nodes using a hashing technique known as consistent hashing (which I will not discuss in detail). The main idea is that a key can be mapped to a set of nodes responsible for it by a simple calculation on the client. ==This means that a client can locate keys without having to query the system for the location of each key==; this saves system resources as hashing is generally faster than performing a remote procedure call.
https://github.com/mixu/vnodehash