Introduction

Relational algebra is a part of the relational model, a mathematical framework for data management invented in the late 1960s. All relational databases (such as PostgreSQL, MySQL, and SQLite) are based on this foundation, and much of their enduring power and popularity derive from these roots. Nevertheless, both internally and externally, relational databases as we know them deviate quite a bit from the relational model.

Internally, they don’t adhere strictly to relational algebra for querying and optimization. Externally, they all use dialects of the SQL language, which lacks many of the qualities of relational algebra.

Bmg is an implementation of relational algebra in Ruby. It can be used to query and transform in-memory datasets, which is very convenient in a range of situations, but not a replacement for a proper database system. Therefore, Bmg can also be used in conjunction with an external SQL database, in which case Bmg functions as a powerful and very flexible interface to the underlying database.

There are two reasons why you might want to learn about relational algebra:

1. To gain a deeper understanding of SQL-based databases and make better use of their features
2. To query and process data with Bmg, using the full power of relational algebra

We will first discuss what exactly algebra and relational means in this context, before showing some non-trivial examples of relational algebra in action.