A very short introduction to relational databases

Relational databases have become ubiquitous in modern-day life, with logins and Airbnb being common uses. For this book, we build our database and interact with the HITRAN database.

A relational database is a structured approach to storing data. It describes a group of tables that are related together. The relationships enable a reduction of data duplication. Typically, these databases are interfaced and queried with using SQL. One of the few programming languages used for about 50 years.

Modern Datatbase applications are oftend distributed over a wide variety of nodes, individual computers, this has a much more complex backend to store partial databases. However, many of these formats have a SQL like interface that can be used to abstract away much of the complexity. Apache Spark is a great example of this.

Note

If you are interested in learning SQL, this textbook is great.

Relational databases reduce duplication by separating out repeated values. For example, If one was to store some data on a customer in a shop, one might store a table as follows,

erDiagram transaction{ transaction_id int coustomer_id int item_id int item_name string price float card_number int loyalty_card bool }

Fig. 2 A table of important information associated with some transaction for a shop.

It should be evident that storing all of these fields for each transaction would be wasteful if a customer regularly bought more than one item. When using a relational database, the number of repeat entries can be cut down by creating multiple tables and linking one table that contains information about a customer to another table that contains information about transactions. This can be described in an entity relationship diagram depicted on the next page.

Note

These two pages are split to avoid a bug with Jupyter-Book.