MongoDB is a NoSQL document database for developers building on JSON-like data, run as the Atlas managed cloud service or self-managed on their own servers.
MongoDB is a NoSQL document database for developers who build applications on JSON-like data rather than fixed tables. It stores records as documents whose fields can differ within one collection, so a change of shape needs no migration. MongoDB, Inc., headquartered in New York and founded in 2007 as 10gen, builds it and owns it outright. The database ships as MongoDB Atlas, a managed cloud service, and as the self-managed Community Edition and Enterprise Advanced. The aggregation pipeline groups and transforms documents in stages such as $group on the cluster itself, without moving data to another platform. Multikey indexes cover array fields, TTL indexes expire documents once a timestamp passes, and multi-document ACID transactions span collections and databases. Sharding spreads a collection across nodes while a background balancer migrates ranges between shards. Change streams publish inserts, updates and deletes as a resumable feed. Read concern and write concern are chosen per operation, from local reads to majority acknowledgement. Text indexes score results by relevance, and on Atlas, Atlas Search extends that while Atlas Vector Search queries embeddings with pre-filters on ordinary fields. Client-side field level encryption encrypts chosen fields inside the application before they cross the network, and continuous cloud backup restores an Atlas cluster to a chosen second. MongoDB suits application developers, platform engineers and data teams, from a single learner to a large enterprise. Atlas fits teams that want the database run for them on AWS, Azure or Google Cloud, from a learning cluster through production workloads. Community Edition fits teams that install and operate the server themselves on local machines or in private data centers. Enterprise Advanced fits organisations that need auditing and Kerberos authentication on self-managed deployments, with search and vector search available as an add-on. Atlas is billed by the hour per cluster, Community Edition is downloaded at no charge and Enterprise Advanced is sold by subscription through sales. In production MongoDB is the operational store behind web and mobile back ends, a search layer over application data and the vector store for AI applications that keep embeddings beside their source documents. The Kafka connector carries change streams into downstream systems. Compass is the desktop GUI for browsing collections and managing indexes, and Atlas Charts draws visualisations from collections through an aggregation query bar. Global Clusters place data in geographic zones, and writes in each zone go to that zone's single primary rather than to several primaries at once. Drivers such as the Java and Rust drivers connect application code to the cluster.
Features
- Included: Schemaless document storage
- Included: Secondary indexes on any field
- Included: Multi-document ACID transactions
- Included: Automatic sharding across nodes
- Included: Tunable consistency per operation
- Not included: Multi-region active-active writes
- Included: Server-side aggregation pipelines
- Included: Change streams for downstream consumers
- Included: Time-to-live expiry on records
- Included: Integrated full-text search on documents
- Included: Vector search on embedded fields
- Not included: Multi-model data access
- Included: Collection-level access roles
- Included: Client-side field-level encryption
- Included: Continuous backup with point-in-time restore
- Not included: Embedded mobile database with sync
- Included: Open source community edition
- Not included: Serverless on-demand capacity pricing
Additional Features
- Atlas managed clusters on AWS, Azure and Google Cloud
- Flexible document model with differing fields in one collection
- Aggregation pipeline stages such as $group run on the cluster
- Multikey indexes over array fields
- TTL indexes with expireAfterSeconds
- Multi-document ACID transactions across collections and databases
- Sharding with a background balancer migrating ranges
- Change streams with resumeAfter and startAfter tokens
Best for
- Developers building applications on JSON-like documents
- Teams that want a managed database run on AWS, Azure or Google Cloud
- AI application teams keeping embeddings beside operational data
- Learners and prototypes on the Free and Flex clusters
- Enterprises self-managing MongoDB in private data centers
- Regulated teams needing collection-level roles and field encryption
- Event-driven architectures feeding Kafka from change streams
- Analysts charting collections without a separate BI stack
Use cases
- Storing records whose fields change without a migration
- Running production clusters on Atlas Dedicated
- Prototyping on a Flex cluster with a monthly cost ceiling
- Expiring session data automatically with TTL indexes
- Streaming inserts and updates into Kafka
- Building semantic search over stored embeddings
- Adding relevance-ranked text search to an application
- Restoring a cluster to a chosen second after a bad write
- Encrypting sensitive fields before they leave the application
- Self-hosting Community Edition in a private data center
- Visualising collections in Atlas Charts
- Sharding a growing collection across nodes
Screenshots & Videos
Explore MongoDB in action