Architecture
editReplacement for mostly-uncited "Main features" section and the previously trimmed "Architecture" section
MongoDB is a document-oriented database that stores data records as BSON (Binary JSON) documents, a binary serialization format that extends JSON-like documents with additional data types. Documents are organized into collections.[1][2][3] Unlike relational databases, which store data in tables with fixed columns, MongoDB collections feature a flexible schema where individual documents can contain different fields and nested structures.[1][2] Documents are sometimes compared to rows in relational databases, although they can represent more complex hierarchical data structures.[2]
Related data can be modeled either by embedding documents within one another to form hierarchical structures, or by using references between documents, depending on application requirements.[1][2] This flexibility allows developers to model data structures that change frequently. Below are two examples of a BSON format with an embedded document:
{
"name":"John Doe",
"address":{
"street":"123 Park Street",
"city":"Anytown",
"state":"NY"
}
}
{
"_id":ObjectId("4b2d75476cc613d5ee930164"),
"username": "Joe",
"hobbies":[
"reading",
"running",
"pickleball"
]
}
MongoDB provides a query language for retrieving, modifying, and processing documents.[1][2] Query operations support filtering, conditional expressions, and index usage to improve query performance.[4] It also includes an aggregation framework for grouping, counting, transforming, and analyzing data.[1][2] An example query to find documents matching a nested object is:
db.people.find({"name":{"first":"Joe", "last":"Schmoe"}})
To maintain redundant copies of data and provide high availability, MongoDB uses a single-primary replica set model rather than multi-master replication. This design simplifies write consistency compared to systems where multiple servers process writes concurrently.[2][4] A replica set contains one primary member that receives write operations and secondary members that asynchronously replicate data from the primary.[5][6] To support horizontal scaling, MongoDB utilizes sharding, which distributes data across multiple servers.[1][2]
MongoDB is implemented in C++.[2][3] It provides command-line interfaces, including the MongoDB Shell, alongside administrative tools for data import, export, and diagnostics.[1][2][4]
Overview
editReplace/combine History & Background sections
Conceived in 2007 by Dwight Merriman, Eliot Horowitz, and Kevin P. Ryan due to their frustrations using structured relational databases in a cloud computing startup, the software was named "MongoDB" from the word humongous.[1][2] Originally released to the public as an open-source project on August 27, 2009,[1][5] it was initially one of several tools developed by a company called 10gen.[1] By 2013, MongoDB had statistically become the most widely-used NoSQL database and the main product of 10gen, prompting the company to change its name to MongoDB Inc.[1][7] The company went public on October 20, 2017.[1][8]
Initially, the software was free and open source under the AGPL license.[9] However, MongoDB adopted the Server Side Public License (SSPL) for all future releases starting in 2018.[10][11] It is available on Windows, macOS, and Linux.[12]
As of May 2025, MongoDB was ranked as the fifth most popular database software globally.[13][14] The platform grew to serve 50,000 customers by 2024.[10] While originally best known as a NoSQL database product,[15] the company launched a database as-a-service cloud platform named MongoDB Atlas in 2016.[9] By 2024, Atlas accounted for 70 percent of MongoDB's total revenue.[10] Over time, the software expanded beyond its initial scope, adding features such as analytics, transactional databases,[16] encryption,[17] vector databases,[10] ACID, migration functionality, and other enterprise tools.[18]
Version history
editReplace current Version History from a comprehensive table to a prose-style summary of independent sources
The first version of MongoDB was written by co-founders Dwight Merriman and Eliot Horowitz around 2007.[19] Version 1 was released on August 27, 2009.[5]: 5 The first version was basic. It could create, read, update, and delete data. It had indexing to improve performance, but no authentication system built-in.[5]: 5 Replication for data backup also existed at release.[5]: 5 Javascript functionality and MapReduce were introduced in version 1.2 in December 2009 and indexing as a background process was added in version 1.4 in March 2010.[5]: 5, 8
Version 2 introduced many of the main features still in the modern MongoDB database.[5]: 8 It was released in September of 2011.[5]: 8 Version 2 introduced sharding and MongoDB's first practical tools for storing geometric data, as well as performance improvements.[5]: 6–8 Version 2 added MongoDB's first built-in authentication functionality, as well as numerous improvements to queries, indexing, security, performance, and an API.[5]: 6–7 In August of 2012, version 2.2 added the ability to set an expiration date on data, as well as MongoDB's aggregation framework, used for complex data operations, and database locking.[1][5]: 8 Version 2.4 added text searches, security improvements and using hashed indexes.[5]: 8 Version 2.6 improved query performance in April 2014 .[5]: 8
In 2014, the WiredTiger storage engine was released, initially as an optional feature, but it became standard in 2014.[1] WiredTiger helped control database locking by preventing one change to a document until another goes through.[5]: 9 Version 3.0 in March 2015 added the ability to swap storage sources.[5]: 8 3.2 in December 2015 allowed users to enforce rules about the structure or data in a document.[5]: 8 Versions 3, 3.2, and 3.4 made several improvements to file storage, sharding, and database views.[5]: 7 In June 2016, MongoDB Atlas was released.[5]: 7 Atlas is MongoDB's database-as-a-service SaaS product.[5]: 296 Version 3.4 added rules for how string data is stored.[5]: 7
In June 2018, version 4 added a set of tools: Mobile, Stitch, Sync, and Kubernetes Operator.[5]: 7–8 These were focused on modern software like internet of things and mobile applications.[5]: 9 It also added retries for writing data and the ability to make numerous database changes as a single transaction, or "multi-document ACID transactions".[5]: 7 This meant MongoDB could be used in applications that have to access up-to-date changes.[5]: 9 Version 4.2, which was released in 2019,[20] improved shards, replicas, and indexing.[5]: 7–8 The biggest improvements were security features like field level encryption.[20] Around the same time, MongoDB released Atlas Data Lake, a tool for managing data on an Amazon AWS S3 server. It also started integrating Realm, a mobile database technology it acquired, into MongoDB.[20]
Version 5 was released in July of 2021.[5]: 8 Version 5 allowed users to change their sharding live and added the Mongosh (also known as "MongoDB Shell") command line interface.[5]: 8 It also marked MongoDB's transition into a cloud-based service.[5]: 9 In version six, released in July 2022, MongoDB began encrypting queries and added Stream Buffer Queue, which improves the performance of queries.[5]: 8 It also added synchronization between different database clusters.[5]: 8 In September 2023, numerous quality of life improvements were introduced for developers.[21] For example, it made it so developers can make changes to an Atlas database on their local machine and introduced Atlas Edge Servers that can be deployed at the data center storing the data. MongoDB also introduced MongoDB Press and expanded its MongoDB University programs.[21] Version 8, released in October of 2024, improved the database's query speed by improving memory usage, making architecture changes, and faster sharding, along with other improvements.[22] Many out-dated features that were rarely used were also removed.[22]
- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 Jones, A. (2024). Comprehensive MongoDB: Advanced Techniques and Best Practices. Walzone Press. Retrieved November 26, 2025.
- 1 2 3 4 5 6 7 8 9 10 11 Hows, David; Plugge, Eelco; Membrey, Peter; Hawkins, Tim (2013-11-04). The Definitive Guide to MongoDB. Berkeley, CA: Apress. ISBN 978-1-4302-5821-6.
- 1 2 "Professional NoSQL | Wiley". Wiley.com. Retrieved 2026-07-13.
- 1 2 3 Chodorow, Kristina; Dirolf, Michael (2011). MongoDB: The Definitive Guide. O'Reilly Media. Retrieved January 28, 2026.
- 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 Giamas, A. (2022). Mastering MongoDB 6.x: Expert techniques to run high-volume and fault-tolerant database solutions using MongoDB 6.x. Packt Publishing. ISBN 978-1-80323-483-0. Retrieved November 26, 2025.
- ↑ Aleksendrić, Marko; Borucki, Arek; Domingues, Leandro; Hammad, Malak Abu; Hannouch, Elie; Nair, Rajesh; Palmer, Rachelle (January 5, 2024). Mastering MongoDB 7.0: Achieve data excellence by unlocking the full potential of MongoDB. Packt Publishing Ltd. ISBN 978-1-83546-319-2. Retrieved January 28, 2026.
- ↑ "10gen embraces what it created, becomes MongoDB Inc". Gigaom. Archived from the original on March 5, 2016. Retrieved January 29, 2016.
- ↑ Witkowski, Wallace (October 21, 2017). "MongoDB shares rally 34% in first day of trading above elevated IPO price". MarketWatch. Dow Jones. Archived from the original on February 26, 2018. Retrieved February 26, 2018.
- 1 2 Sawers, Paul (October 10, 2018). "MongoDB to acquire cloud database provider MLab for $68 million". VentureBeat. Retrieved August 10, 2024.
- 1 2 3 4 Sawers, Paul (April 28, 2024). "MongoDB CEO Dev Ittycheria talks AI hype and the database evolution as he crosses 10-year mark". TechCrunch. Retrieved August 11, 2024.
- ↑ Lardinois, Frederic (October 16, 2018). "MongoDB switches up its open-source license". TechCrunch. Retrieved August 10, 2024.
- ↑ Jones, Adam (January 8, 2025). Applied MongoDB: Strategies for Designing and Developing Robust NoSQL Databases. Walzone Press. Retrieved January 28, 2026.
- ↑ Kanaracus, Chris (January 11, 2019). "AWS, MongoDB database collision stirs open source tensions". SearchAWS. Retrieved August 10, 2024.
- ↑ "DB-Engines Ranking". DB-Engines. January 1, 2024. Retrieved August 27, 2024.
- ↑ Lardinois, Frederic (March 7, 2017). "MongoDB's Atlas database service goes freemium". TechCrunch. Retrieved August 10, 2024.
- ↑ Baer, Tony (June 21, 2022). "Are Snowflake and MongoDB on a collision course?". VentureBeat. Retrieved August 11, 2024.
- ↑ Newman, Lily Hay (June 7, 2022). "A Long-Awaited Defense Against Data Leaks May Have Just Arrived". Wired. Retrieved August 11, 2024.
- ↑ Brust, Andrew (June 7, 2022). "MongoDB fires up new cloud, on-premises releases". VentureBeat. Retrieved August 11, 2024.
- ↑ Bort, Julie (October 20, 2017). "People told the MongoDB founders they were 'completely crazy' and now the company is worth $1.6 billion". Business Insider. Retrieved December 10, 2025.
- 1 2 3 Lardinois, Frederic (June 18, 2019). "MongoDB gets a data lake, new security features and more". TechCrunch. Retrieved December 10, 2025.
- 1 2 Barron, Jenna (September 26, 2023). "MongoDB announces new database capabilities to improve developer experience". SD Times. Retrieved February 25, 2026.
- 1 2 Jackson, Joab (October 2, 2024). "MongoDB 8 Goes Hard on Time-Series Data, Horizontal Scaling". The New Stack. Retrieved February 25, 2026.