Skip to content

The 'M' in MERN: Introduction to MongoDB

Every time we turn off our Node.js server, the variables holding our arrays (like the fakeDatabase in our MVC chapter) are immediately securely erased from the computer's temporary memory (RAM). When we restart the server, the data is gone forever.

To build an application like Instagram, Amazon, or Spotify, we clearly cannot rely on temporary memory. We need a permanent, highly searchable, secure filing system on a hard drive. We need a Database.

In this chapter, we explore the "M" in the MERN stack: MongoDB.

SQL vs. NoSQL (Why MongoDB?)

For decades, the undisputed king of databases was the Relational Database (SQL databases like MySQL or PostgreSQL). SQL databases are essentially massive Excel spreadsheets. They use rigid Tables consisting of strictly predefined columns and rows.

While SQL is incredibly powerful, it has a steep learning curve and uses a completely different syntax than JavaScript. Furthermore, if you suddenly decide that users in your application can have multiple shipping addresses instead of just one, altering a rigid SQL table structure can be a massive migraine.

MongoDB is a NoSQL Database. Instead of rigid Tables and Rows, MongoDB embraces flexibility. What makes MongoDB the absolute perfect companion for a Node.js backend is that it inherently speaks JavaScript! MongoDB stores data in a format called BSON (Binary JSON). To a MERN developer, interacting with the database feels exactly like manipulating standard JavaScript objects.

Translating the Terminology

If you have any prior experience with SQL (Relational) databases, here is a quick translation guide to MongoDB's NoSQL terminology:

graph LR
 subgraph SQL World
 DB1[(Database)] --> T[Table]
 T --> R[Row]
 T --> C[Column]
 end

 subgraph MongoDB World
 DB2[(Database)] --> Col[Collection]
 Col --> Doc[Document]
 Col --> F[Field]
 end
  1. Database: Same concept. The overarching container for your application's data.
  2. Collection: Equivalent to an SQL Table. You will have a users collection, a products collection, and a reviews collection.
  3. Document: Equivalent to an SQL Row. A single individual user profile or a single product is a Document.
  4. Field: Equivalent to an SQL Column. The specific keys inside the Document (like email: "admin@test.com").

The Anatomy of a MongoDB Document

Let's look at what a Document actually looks like when stored inside a MongoDB Collection. It looks remarkably identical to a JavaScript Object!

{
 "_id": "64a7f05c3b9d140e9c8a4d7a",
 "username": "MernNinja",
 "email": "ninja@mern.com",
 "age": 28,
 "isAdmin": true,
 "hobbies": ["Coding", "Gaming", "Pizza"],
 "address": {
 "city": "New York",
 "zipcode": "10001"
 },
 "createdAt": "2024-05-12T08:30:00Z"
}

The Magic of the _id Field

Notice the very first field: _id. In SQL, you have to manually configure auto-incrementing primary keys (1, 2, 3...).

MongoDB automatically handles this for you. Whenever you insert a new Document into a Collection, MongoDB instantaneously generates a massive, universally unique 24-character hexadecimal String called an ObjectId and assigns it to the _id field. You use this _id to quickly and easily search for, update, or delete the Document later.

Flexibility and Nesting

Notice the hobbies field is a true Array, and the address field is a deeply nested Object. SQL databases despise arrays; you would normally have to create an entirely separate Table just to hold hobbies and meticulously link them to the user. MongoDB simply embeds the array directly into the Document. This nested structure makes data retrieval incredibly fast and intuitive.

Where Does MongoDB Live? (MongoDB Atlas)

Historically, you had to install MongoDB directly onto your laptop, run a complex background service, and manage the hardware yourself.

Today, the standard practice in the industry is to use MongoDB Atlas.

Atlas is a fully managed cloud database service hosted by the creators of MongoDB. Instead of running the database locally, MongoDB provides you with a highly secure cluster hosted on Amazon Web Services (AWS), Google Cloud, or Microsoft Azure.

When you set up an Atlas cluster, they give you a Connection String (a URI) that looks something like this:

mongodb+srv://admin_user:SuperSecretPassword123@cluster0.mongodb.net/MyApplicationDB?retryWrites=true&w=majority

Your Node.js server uses this String as a secret password to securely reach out across the internet, connect to the Atlas cluster, and begin reading and writing data safely.

Summary

In this introduction, we explored why MongoDB dominates the modern Node.js ecosystem: 1. NoSQL vs SQL: We traded rigid Tables and Rows for flexible, JSON-like data storage. 2. Structural Terminology: Databases contain Collections (like tables), which contain numerous Documents (like individual rows). 3. The Document Advantage: MongoDB stores nested arrays, sub-objects, booleans, and strings natively, and effortlessly assigns a unique _id primary key to every Document automatically. 4. MongoDB Atlas: The industry standard for hosting your database safely in the cloud, accessed via a secure Connection String URI.

But how do we establish that connection? How do we tell our Express Controller to send a newly registered user directly into a MongoDB Collection? In the next chapter, we introduce the elegant bridge that connects Node.js to MongoDB: Mongoose.