Definition
NoSQL stands for "Not Only SQL" and refers to a category of databases designed to store and retrieve data in flexible, non-relational formats. Unlike traditional relational databases, which use structured tables and schemas, NoSQL databases allow unstructured, semi-structured, or structured data to be stored in formats such as documents, key-value pairs, graphs, or wide columns.
This flexibility makes NoSQL databases well-suited for applications with large volumes of data, rapid scalability needs, or varied data structures. They are commonly used in real-time applications, big data processing, and cloud-native systems.
Advanced
At an advanced level, NoSQL databases are divided into several types, including document databases such as MongoDB, key-value stores such as Redis, graph databases such as Neo4j, and wide-column stores such as Cassandra. They are often designed for distributed computing, allowing data replication and sharding across clusters for high availability and horizontal scaling.
While NoSQL offers flexibility, it typically sacrifices some ACID (Atomicity, Consistency, Isolation, Durability) guarantees in favour of eventual consistency. Advanced implementations may combine NoSQL with relational systems in polyglot architectures for maximum efficiency.
Why it matters
- Handles large volumes of structured and unstructured data.
- Provides horizontal scalability for cloud and distributed systems.
- Enables faster development by reducing schema constraints.
- Powers real-time and high-performance applications.
Use cases
- Storing user profiles and activity data in social media apps.
- Managing product catalogues in e-commerce platforms.
- Powering recommendation engines with graph databases.
- Handling session storage and caching with key-value stores.
Metrics
- Query response time under high load.
- Data replication and consistency performance.
- Horizontal scaling efficiency across clusters.
- Cost savings in infrastructure compared to relational databases.
Issues
- Eventual consistency may not suit mission-critical data.
- Less standardised query languages compared to SQL.
- Complex transactions can be harder to manage.
- Requires specialised knowledge for scaling and optimisation.
Example
A global e-commerce platform uses MongoDB, a document-oriented NoSQL database, to manage its dynamic product catalog. The system allows new attributes to be added without altering rigid schemas. This flexibility reduces development time and supports rapid scaling during seasonal sales.