SQL = Structured Query Language or NoSQL?

SQL-- allows you to write queries, not a database itself.

SQL is for Relational DBs

(example: Postgres)

• Data is stored in clear schema relations; tables
• Numbered rows of clearly defined data
• Great for dynamic data that is used on different parts of app a lot; heavy-duty transactional traffic
• Allows DB query for info: SELECT category FROM table;
• Vertically scalable by increasing horsepower of hardware

NoSQL is for Distributed, non-relational DBs

(example: MongoDB)

• Data is stored in "collections" of JSON-looking docs; docs have no uniform structure and no relations unless manually implemented. dynamic schema for unstructured data
• Key-value pairs, graph databases or wide-valued stores
• Great for data that is read a lot but not updated a lot--static?
• Great for heavily used features that display lots of data--fast.
• Horizontally scalable by increasing the number of machines

sources:

the geek stuff: SQL vs NoSQL

highly scalable blog: noSQL data modeling techniques