NoSQL Databases

Amazon DocumentDB vs Azure Cosmos DB

Compare Amazon DocumentDB and Azure Cosmos DB side-by-side. See feature differences and pricing in the NoSQL Databases category.

Amazon DocumentDB

by Amazon

No reviews yet

Pricing: Usage-based, US East (Ohio): Serverless $0.0822 per DCU-hour, db.r8g.large instance $0.276 an hour, Elastic Clusters $0.132 per vCPU-hour, plus storage and I/O; one-month trial

Amazon DocumentDB is AWS's managed document database for apps built on MongoDB APIs and drivers, run on serverless capacity, instances or sharded clusters.

13 /18 features Full profile

Azure Cosmos DB

by Microsoft

No reviews yet

Pricing: Free tier with 1,000 RU/s and 25 GB a month. Provisioned throughput $5.84 per 100 RU/s a month, autoscale $8.76; serverless billed per request unit; storage extra.

Azure Cosmos DB is Microsoft's managed NoSQL database on Azure, with NoSQL, MongoDB, Cassandra, Gremlin and Table APIs, for global and AI apps, billed by usage.

16 /18 features Full profile

Feature Comparison

NoSQL Databases features side-by-side

Feature	Amazon DocumentDB 13/18 features	Azure Cosmos DB 16/18 features
Schemaless document storage Records in one collection can carry different fields with no predefined table definition or migration.	Yes	Yes
Secondary indexes on any field Indexes on nested fields or array elements, not only on the primary key or partition key.	Yes	Yes
Multi-document ACID transactions Writes touching several documents, rows or collections commit or roll back together.	Yes	Yes
Automatic sharding across nodes Data partitioned and rebalanced across servers by a shard key without manual resharding.	Yes	Yes
Tunable consistency per operation Read and write consistency chosen per request, from eventual to majority or strong.	Yes	Yes
Multi-region active-active writes Writes accepted in more than one geographic region with conflict handling between them.	No	Yes
Server-side aggregation pipelines Group, join and transform documents inside the database instead of in application code.	Yes	Yes
Change streams for downstream consumers Applications subscribe to inserts, updates and deletes as an ordered, resumable event feed.	Yes	Yes
Time-to-live expiry on records Documents or rows are deleted automatically once a per-record expiry timestamp passes.	Yes	Yes
Integrated full-text search on documents Text indexes with relevance ranking built into the database, with no separate search cluster to run.	Yes	Yes
Vector search on embedded fields Embeddings stored on documents and queried by nearest neighbor alongside ordinary filters.	Yes	Yes
Multi-model data access The same engine serves documents plus key-value, graph or wide-column access patterns.	No	Yes
Collection-level access roles Permissions granted per database and per collection or table to named roles.	Yes	Yes
Client-side field-level encryption Selected fields are encrypted by the driver before leaving the application and are unreadable by the server.	Yes	Yes
Continuous backup with point-in-time restore Oplog or commit-log backups let you restore a cluster to any second within the retention window.	Yes	Yes
Embedded mobile database with sync A device-side database that syncs with the server cluster for offline-first mobile apps.	No	No
Open source community edition A community edition whose source is public and which can be self-hosted without a commercial license.	No	No
Serverless on-demand capacity pricing Capacity billed per request or per operation, with no instance to size in advance.	No	Yes
Overall Rating	No reviews yet	No reviews yet

Amazon DocumentDB 13/18 features

Schemaless document storage: Yes
Secondary indexes on any field: Yes
Multi-document ACID transactions: Yes
Automatic sharding across nodes: Yes
Tunable consistency per operation: Yes
Multi-region active-active writes: No
Server-side aggregation pipelines: Yes
Change streams for downstream consumers: Yes
Time-to-live expiry on records: Yes
Integrated full-text search on documents: Yes
Vector search on embedded fields: Yes
Multi-model data access: No
Collection-level access roles: Yes
Client-side field-level encryption: Yes
Continuous backup with point-in-time restore: Yes
Embedded mobile database with sync: No
Open source community edition: No
Serverless on-demand capacity pricing: No

No reviews yet

Azure Cosmos DB 16/18 features

Schemaless document storage: Yes
Secondary indexes on any field: Yes
Multi-document ACID transactions: Yes
Automatic sharding across nodes: Yes
Tunable consistency per operation: Yes
Multi-region active-active writes: Yes
Server-side aggregation pipelines: Yes
Change streams for downstream consumers: Yes
Time-to-live expiry on records: Yes
Integrated full-text search on documents: Yes
Vector search on embedded fields: Yes
Multi-model data access: Yes
Collection-level access roles: Yes
Client-side field-level encryption: Yes
Continuous backup with point-in-time restore: Yes
Embedded mobile database with sync: No
Open source community edition: No
Serverless on-demand capacity pricing: Yes

No reviews yet

Pricing

Amazon DocumentDB

Usage-based, US East (Ohio): Serverless $0.0822 per DCU-hour, db.r8g.large instance $0.276 an hour, Elastic Clusters $0.132 per vCPU-hour, plus storage and I/O; one-month trial

See all plans

Azure Cosmos DB

Free tier with 1,000 RU/s and 25 GB a month. Provisioned throughput $5.84 per 100 RU/s a month, autoscale $8.76; serverless billed per request unit; storage extra.

See all plans

Explore Each Product

Amazon DocumentDB Read reviews, features & pricing

Azure Cosmos DB Read reviews, features & pricing

Looking for more options?

Browse all NoSQL Databases