Working mostly with relational databases it is fascinating and eyes-opening to learn about DynamoDB Single Table Design.
DynamoDB tables consist of keys and attributes. What is interesting is that keys and attributes do not have to be homogenous, means it is possible to store absolutely different records in the same DB and sucessfully operate that structure.
Primary key - is a partition key OR partition + sort keys
Partition key - a top-level primary identifier. The data would be split by that key
Sort key - used for sorting and querying. Could be composed from multiple values and queried by prefixes.
Secondary index (local, global) - allows additional querying scenarios. Local - with the same partition key as the original table.
Single Table Design
The main reason for using a single table in DynamoDB is to retrieve multiple, heterogenous item types using a single request.
Links