MongoDB Indexing 101: Stop the Collection Scans

Welcome to Week 2! 🚀 This week is all about Speed.

You can have the best schema in the world (from Week 1), but if you don’t index your data, your app will crash under load.

1. The “Collection Scan” (The Villain)

Imagine looking for “Waldo” in a book, but you have no index page. You have to flip through every single page until you find him. In MongoDB, this is called a COLLSCAN (Collection Scan).

If you have 1 million users and you run: db.users.find({ email: "honey@example.com" })

Without an index, MongoDB reads 1,000,000 documents to find the one matching email. Ouch.

2. What is an Index? (The Hero)

An index is a small, sorted data structure (B-Tree) that stores a small portion of the collection’s data.

If you index email: db.users.createIndex({ email: 1 })

MongoDB creates a sorted list of emails pointing to the documents. Now, finding ”honey@example.com” is like using a phonebook. It takes logarithmic time (O(log N)). Instead of 1 million reads, it might take just 3 or 4 ops.

3. Types of Indexes

A. Single Field Index

The most basic type. Indexes one field.

// Index on 'username' in ascending order (1)
db.users.createIndex({ username: 1 })

Note: Sorting order (1 vs -1) doesn’t matter much for single field indexes because MongoDB can traverse the index in either direction.

B. Unique Index

Enforces uniqueness. Good for emails, usernames, slugs.

db.users.createIndex({ email: 1 }, { unique: true })
// Now, inserting a duplicate email throws an error!

C. Default Index (_id)

MongoDB automatically indexes the _id field. You can’t delete this one.

4. The Cost of Indexes

“Why not index everything?” 🤔

1. Write Penalty: Every time you Insert, Update, or Delete a document, MongoDB must update the index too.
2. Storage Cost: Indexes take up RAM and Disk space.

Rule of Thumb: Index fields that you Query or Sort by frequently.

5. Analyzing a Query

How do you know if an index is being used? Use .explain("executionStats").

db.users.find({ username: "honeycoder" }).explain("executionStats")

Look for:

• COLLSCAN: Bad! No index used.
• IXSCAN: Good! Index scan used.
• totalDocsExamined: Should match the number of results (ideally 1).

🧠 Daily Challenge

1. Open your project/terminal.
2. Pick a collection with some data (or generate dummy data).
3. Run a query with .explain("executionStats") and see the COLLSCAN.
4. Create an index on that field.
5. Run explain again and see the IXSCAN and the huge drop in totalDocsExamined.

See you on Day 2, where we level up to Compound Indexes! ⚡

Next Step

Next: Compound And Multikey Indexes →