A NoSQL database stores and retrieves data without the fixed tables and relations of traditional SQL systems. It uses documents, key-value pairs, wide columns or graphs, and is often built to scale horizontally.
SQL (Structured Query Language) is the standard language for defining, querying and changing data in relational databases. It is used to filter, join, aggregate and update rows stored in tables.
The difference is the data model: SQL databases keep related tables under a fixed schema, and NoSQL databases use flexible documents, key-value pairs or graphs. SQL is strongest at joins and transactions; NoSQL often scales out more easily.
A vector database stores embeddings and quickly finds the ones most similar to a query vector, usually alongside metadata filters. It is the common retrieval layer behind semantic search, recommendations and RAG.
Caching stores copies of frequently used data in a faster location so future requests can be served without repeating slow work. That slow work might be a database query, a computation or a network call.
A time-series database is built to store and query measurements indexed by time, such as metrics, sensor readings or prices, at very high write volume.
A clear proven process, that delivers
You've seen our ideas - now go deeper.
Discover more insights, guides, and engineering practice.