Filter + project essentials
Write a query to return active users only, showing name and email, sorted by newest signup date.
Data / Knowledge library
MongoDB
document-oriented nosql database for flexible apps
Why learn MongoDB?
MongoDB helps teams store and query semi-structured data quickly without rigid table schemas, making it easier to ship features when data shapes evolve.
Where it came from
2009: MongoDB was first released by 10gen as a document database. 2013: 10gen rebranded as MongoDB, Inc. 2018: MongoDB 4.0 introduced multi-document ACID transactions (replica sets). 2019: MongoDB 4.2 extended transactions to sharded clusters. 2018 onward: SSPL licensing and rapid growth of Atlas as the managed default.
- 2009: MongoDB was first released by 10gen as a document database.
- 2013: 10gen rebranded as MongoDB, Inc.
- 2018: MongoDB 4.0 introduced multi-document ACID transactions (replica sets).
- 2019: MongoDB 4.2 extended transactions to sharded clusters.
- 2018 onward: SSPL licensing and rapid growth of Atlas as the managed default.
Build your understanding
Work through these exercises in order. Try the task before opening a reference.
- Write a query to return active users only, showing name and email, sorted by newest signup date.
- For an e-commerce app, decide what to embed vs reference for orders, line items, and products. Justify in 4 bullet points.
- Build an aggregation that returns total revenue per category for the last 30 days, then add an index plan to speed it up.
- Create a collection validator that requires `email` and `createdAt`, and restricts `status` to `active|inactive`.
- Given a slow query with `COLLSCAN`, identify one missing index and one query rewrite.
More ways to practice
Modeling tradeoff drill
For an e-commerce app, decide what to embed vs reference for orders, line items, and products. Justify in 4 bullet points.
Aggregation mini-report
Build an aggregation that returns total revenue per category for the last 30 days, then add an index plan to speed it up.
Schema validation
Create a collection validator that requires `email` and `createdAt`, and restricts `status` to `active|inactive`.
Error spotting
Given a slow query with `COLLSCAN`, identify one missing index and one query rewrite.
Documentation and sources
- MongoDB Manual (www.mongodb.com)
- MongoDB Atlas Docs (www.mongodb.com)
- MongoDB Node.js Driver (www.mongodb.com)
- MongoDB University (free courses) (learn.mongodb.com)
Catalog research notes
- MongoDB 8.0 emphasized faster reads/writes and better overall performance.
- Atlas expanded vector search and AI-app support for semantic and hybrid retrieval.
- Queryable Encryption support broadened, improving security for sensitive fields.
- Atlas continued improving real-time data tooling (search, streams, analytics integration).
Connect the ideas
Alternatives
Explore next
Real World Example
Dynamic Inventory Schema for Global Retailers
Global e-commerce retailers managing diverse product catalogs with rapidly changing attributes.
The business problem
Relational databases force rigid schemas that require expensive migrations whenever a new product category, such as electronics with technical specs or apparel with sizing variants, is introduced to the catalog.
How this topic solves it
MongoDB uses a flexible document data model that allows developers to store non-uniform product data in a single collection, mapping directly to application objects without complex object-relational mapping.
- Define a base product document structure containing common fields like SKU and price.
- Embed category-specific attributes as nested documents or arrays within the product document.
- Insert documents into a collection without requiring a pre-defined schema for every variation.
- Utilize the query engine to retrieve products based on nested attribute values.
- Scale the catalog horizontally using sharding to distribute product data across multiple servers.
This diagram illustrates how a central product collection handles diverse data structures through embedding.
Central storage for all inventory items
Contains technical specs array
Contains size and fabric nested objects
Maps JSON directly to code objects
- Product Collection → contains → Electronics Document
- Product Collection → contains → Apparel Document
- Application Layer → reads/writes → Product Collection
Implementation example
This demonstrates how a document can store unique fields like 'specs' without needing a separate table for electronics.
db.products.insertOne({ sku: 'ELEC-001', name: 'Laptop', specs: { ram: '16GB', cpu: 'Octa-core' } }); // Embeds specific attributes directlyExpected business outcome
Developers can iterate on product catalog features and support new inventory types without downtime or schema migration overhead.
Limitations and checks
The flexibility of dynamic schemas can lead to data inconsistency if application-level validation is not strictly enforced; check for this by auditing document structures periodically.