Azure Cosmos DB
By Microsoft
Azure Cosmos DB is Microsoft's fully managed, globally distributed Multi-Model Database service that supports multiple data models and APIs — including document, key-value, graph, and column-family — over a single underlying engine.
Definition
Azure Cosmos DB is Microsoft's fully managed, globally distributed Multi-Model Database service that supports multiple data models and APIs — including document, key-value, graph, and column-family — over a single underlying engine.
Overview
Cosmos DB's defining feature is its multi-API surface: the same globally distributed backend can be accessed through a native SQL/Core API for JSON documents, a MongoDB-compatible API, a Cassandra-compatible API, a Gremlin API for graph data, and a Table API — letting teams pick the interface that matches their application without switching the underlying infrastructure. It offers tunable consistency levels ranging from strong to eventual, letting developers trade off latency and consistency per application, and provides automatic, transparent Database Replication and failover across any number of Azure regions with single-digit-millisecond read and write latency guarantees at the 99th percentile. Cosmos DB competes with other globally distributed, multi-model or multi-API cloud databases such as Google Cloud Spanner for global relational-style consistency and Amazon DynamoDB for key-value/document workloads, positioning itself as Microsoft's answer to teams needing planet-scale, low-latency data access.
Key Features
- Multiple APIs over one engine: SQL/Core, MongoDB, Cassandra, Gremlin, Table
- Tunable consistency levels from strong to eventual
- Automatic multi-region replication and failover
- Single-digit-millisecond latency SLAs at the 99th percentile
- Automatic and transparent horizontal partitioning
- Serverless and provisioned throughput pricing options