Database
Definition
An organized collection of information stored electronically, functioning like a digital filing cabinet with fast search capabilities.
Use Cases
- Netflix: Storing and serving user personalization data and application state at global scale — Netflix uses Apache Cassandra (a distributed NoSQL database) as a core datastore for high-throughput, highly available workloads across multiple regions. (Improved resilience and uptime for customer-facing services by using a database designed for horizontal scaling and fault tolerance.)
- Airbnb: Managing transactional data for bookings, users, and payments — Airbnb has used MySQL as a primary relational database for core transactions, with replication and operational practices to support reliability and growth. (Supports consistent, ACID transactions needed for reservations and payments while scaling operations as the business grew.)
- Stack Overflow: Storing questions, answers, comments, votes, and user accounts — Stack Overflow is known for using Microsoft SQL Server as its primary relational database, with performance tuning and indexing to support heavy read/write traffic. (Fast query performance for search, sorting, and page loads, enabling a responsive experience for a large developer community.)
Provider Equivalents
- AWS: Amazon RDS
- Azure: Azure SQL Database
- GCP: Cloud SQL
- OCI: Oracle Autonomous Database
Frequently Asked Questions
- What's the difference between a database and a data warehouse?
- A database is built for day-to-day application work (creating, updating, and reading records quickly), often called OLTP. A data warehouse is optimized for analytics (large scans, aggregations, reporting), often called OLAP, and usually stores historical data from many sources.
- When should I use a database?
- Use a database when your application needs to store and retrieve data reliably—such as user accounts, orders, inventory, messages, or logs you need to query. Choose a relational database when you need strong consistency and transactions (e.g., payments). Consider a NoSQL database when you need flexible schemas, very high scale, or low-latency access patterns (e.g., user sessions, event data).
- How much does a database cost in the cloud?
- Cost depends on the database type (relational vs NoSQL), deployment model (managed service vs self-managed), instance size/CPU/RAM, storage size and type (SSD vs HDD), read/write IOPS, backup storage, data transfer, and high availability (multi-zone/replicas). Many providers offer free tiers or small starter instances, but production costs typically grow with performance, storage, and availability requirements.
Category: data
Difficulty: basic
Related Terms
See Also