Nosql

TL;DR

A class of databases (short for “Not Only SQL”) that store and manage data using multiple models instead of relational tables.
Suited for large-scale data, datasets with variable structure, and complex relationships; often designed for easy horizontal scaling.
Typically faster for large distributed workloads but generally less mature and less suitable for complex queries and transactions compared with traditional SQL databases.

Definition

NoSQL, or “Not Only SQL,” is a type of database management system that is designed to handle large amounts of data that may not be structured in a traditional, relational way. Unlike traditional SQL databases, which are based on the structured query language (SQL) and use tables to organize data, NoSQL databases use a variety of approaches to store and manage data, including document-based, key-value, column-based, and graph-based systems.

Explanation

NoSQL databases adopt several storage models (document, key-value, column, graph) rather than the fixed table-and-row schema of relational databases. This flexibility lets them store data with variable structures and nested or complex relationships. Document-based systems store data as documents containing any number of fields, including text, numbers, arrays, and nested structures. Column-based systems store data by columns rather than rows, with each column able to contain different data types.

NoSQL systems are commonly designed for distributed deployments, allowing data to be stored across multiple servers to improve performance and reliability. They are often chosen for their scalability and performance when handling large volumes of data. However, NoSQL databases are generally less mature than traditional SQL databases and may lack comparable feature sets. They can also be less well-suited for complex queries and multi-operation transactions that are common in relational systems.

Examples

MongoDB (document-based)

MongoDB is a document-based NoSQL database. In this model, data is stored as documents that are similar to rows in a relational database but with a more flexible structure. Each document can contain any number of fields—including text, numbers, and arrays—and can be nested to represent complex relationships. This flexibility allows MongoDB to store and manage large amounts of data that are difficult to represent in a traditional SQL database, such as data with variable structures or complex relationships.

Cassandra (column-based)

Cassandra is a column-based NoSQL database. In a column-based database, data is stored in columns rather than rows, and each column can contain different data types. Cassandra is designed to handle large amounts of data and is often used in distributed systems, where data is stored across multiple servers to improve performance and reliability. Cassandra is also designed to be highly scalable, allowing it to handle large amounts of data as it grows over time.

Use cases

Handling large amounts of data that outgrow single-server solutions.
Storing data with variable structures that do not fit a fixed relational schema.
Representing complex or nested relationships within documents.
Distributed systems where data is spread across multiple servers for performance and reliability.
Systems that require easy horizontal scalability as data grows.

Notes or pitfalls

NoSQL databases are generally less mature than traditional SQL databases and may lack comparable features and functionality.
They may not be as well-suited for complex queries and transactions typical of relational databases.

SQL
Document-based
Key-value
Column-based
Graph-based
MongoDB
Cassandra