Documentstores

Document stores, or document-oriented databases, are a class of NoSQL databases designed to store, retrieve, and manage collections of documents. A document is a self-contained data unit, typically represented as JSON, BSON, or XML, with a unique identifier. Documents in a collection may have varying structures, enabling schema flexibility.

Key features include indexing of document fields, support for nested objects and arrays, and expressive query languages that allow filtering, projection, and aggregation. Many document stores also provide full-text search, geospatial queries, and built-in replication, sharding, and durability options.

Storage architecture and transactions: Documents are stored in collections. Some systems support ACID transactions at the document level or across multiple documents in newer versions. Consistency models vary; many aim for eventual consistency with tunable durability.

Data modeling: Designers often denormalize and embed related data within a single document, or store references to other documents. This reduces the need for joins and can improve read performance but may lead to data duplication and more complex updates.

Use cases include content management, product catalogs, user profiles, session stores, and mobile or web applications with evolving schemas, as well as logging and event data.

Prominent examples include MongoDB, Couchbase, CouchDB, RavenDB, ArangoDB, and Amazon DocumentDB (MongoDB-compatible). Each offers different query languages, consistency guarantees, and deployment options, from on-premises to cloud-managed services.