Key Concepts

Document-oriented databases are a type of NoSQL database that store data in a document format, typically using JSON or BSON. Unlike traditional relational databases, document-oriented databases do not require a fixed schema, allowing for more flexibility in data modeling. Here are some key concepts to understand when working with document-oriented databases:

Documents

Documents are the basic unit of data in a document-oriented database. They are similar to rows in a relational database but with a few key differences. Documents are typically stored in a format such as JSON or BSON, which allows for nested data structures and more flexible data modeling. Unlike rows in a relational database with fixed columns, documents can also have different fields.

For example, consider a document representing a user in a social media application. This document might include fields such as the user’s name, age, address, and friends list. The friends’ field could contain an array of documents, each representing a friend and containing their name, age, and other information.

{
	"name": "John Smith",
	"age": 30,
	"address": {
		"street": "123 Main St",
		"city": "Anytown",
		"state": "CA",
		"zip": "12345"
	},
	"friends": [
			{
				"name": "Jane Doe",
				"age": 28,
				"address": {
				"street": "456 Elm St",
				"city": "Anytown",
				"state": "CA",
				"zip": "12345"
			},
			{
				"name": "Bob Johnson",
				"age": 32,
				"address": {
				"street": "789 Oak St",
				"city": "Anytown",
				"state": "CA",
				"zip": "12345"
			}
	]
}

Collections

Collections are similar to tables in a relational database. They are a grouping of related documents and can have different fields and data types. Collections can also have indexes to improve query performance.

For example, consider a collection of users in a social media application. This collection might include documents representing each user and indexes on fields such as the user’s name or age to improve query performance.

db.createCollection('users');
db.users.createIndex({ name: 1 });
db.users.createIndex({ age: 1 });

Queries

Queries in document-oriented databases are typically done using a query language such as MongoDB’s query language. Queries can be used to filter documents based on specific criteria, such as field values or nested data structures.

For example, consider a query to find all users in a social media application who are over the age of 25.

db.users.find({ age: { $gt: 25 } });

This query would return all documents in the user’s collection with an age field greater than 25.

Aggregation

Aggregation is the process of performing calculations on a set of documents in a collection. This can include grouping documents by a specific field, calculating averages or sums, and more.

For example, consider an aggregation to count the number of users in a social media application by state.

db.users.aggregate([
  {
    $group: {
      _id: '$address.state',
      count: {
        $sum: 1,
      },
    },
  },
]);

This aggregation would group all documents in the users’ collection by the state field in the address subdocument and then count the number of documents in each group.

Key Concepts

Documents

Collections

Queries

Aggregation

Further Readings