The article in general is very good, but you seem to have confused Dynamo (from Amazon) with DynamoDB (from Amazon). The Dynamo paper describes a different database (node based, nice discovery mechanism) than DynamoDB the scalable, provisioned qps KV store. You might want to correct the name throughout the article, and be careful about using the Dynamo paper (inspiring databases with designs like Cassandra, Riak) to infer anything about the structure or implementation of DynamoDB. This is a common mistake and many people do not realize they are very different implementations. The quote from wikipedia “Amazon DynamoDB is “built on the principles of Dynamo”” is far too generous.