MongoDB Database Guide

What is a MongoDB Database?

MongoDB is an open-source, flexible NoSQL database management system.

MongoDB database stores data in flexible binary JSON (BSON). This schema design suits it for handling multiple collections of unstructured or semi-structured data, making MongoDB a popular choice in modern web development, particularly for big data applications.

MongoDB offers a free cloud platform and downloadable upgrades (enterprise and community editions).

Cloud MongoDB Atlas. Also known as MongoDB database-as-a-service, is the cloud-based version. MongoDB databases are composed of a sandbox for prototyping, storage, shared RAM, and some capacity for scaling.

Community and Enterprise versions. Both of these versions may be hosted on the MongoDB server and are also available for download and self-hosting. The Community and Enterprise versions of the MongoDB distributed database offer the power and resilience of multiple servers that are production-grade and developer-ready. MongoDB Enterprise also offers some features not available in the MongoDB Community version, such as auditing, an in-memory storage engine, encryption for data at rest, and Kerberos access controls and authentication.

How to Use MongoDB Database

MongoDB is especially popular for big data applications and other use cases for semi-structured and unstructured data. Whatever the application, manage MongoDB databases based on their core characteristics:
  • Document model. MongoDB stores data in flexible documents, and each can have its own structure. This allows for easy representation of complex data types.
  • Collections and documents. Documents are organized into collections (similar to tables in relational databases). However, unlike tables, collections do not enforce a schema across all documents. This flexibility allows for easier adaptability to changing data requirements.
  • MongoDB query language. Based on JavaScript, this allows users to retrieve and manipulate data.
  • Supported indexing on any field. This makes data retrieval more efficient and optimizes query performance down to the document level.
  • Aggregation framework. MongoDB provides a framework for complex data aggregation operations, such as grouping, sorting, and filtering, directly within the database.

How to Use MongoDB Database Tools

MongoDB database tools help developers and administrators manage MongoDB databases efficiently. Some of the basic MongoDB database tools include:
  • Mongo/MongoDB shell. This interactive JavaScript interface for MongoDB allows users to interact directly with the database using commands. It can be used for querying, updating, and managing MongoDB databases.
  • Mongod. The primary daemon process for the MongoDB server manages data requests, performs data storage operations, and handles client connections.
  • Mongos/MongoDB shard router. This is a process that acts as a proxy between the application and the sharded cluster and routes queries and write operations to the appropriate shards.
  • Mongodump and mongorestore. These backup and restore MongoDB databases.
  • Mongoimport and mongoexport. These enable the import and export of various formats of data (such as CSV, JSON, or BSON).
  • Replica sets. Groups of replica sets or multiple MongoDB instances—one primary key and one or more secondary nodes—maintain the same data to provide redundancy and high availability, and minimize data loss. The primary key also uniquely identifies documents within a collection, further optimizing data retrieval and management.

How to Connect Databases in MongoDB

There are a few decisions en route to getting MongoDB up and running, and how to connect to MongoDB databases depends in large part on which methods you select. The most common approach is through a MongoDB client or driver.

First, install MongoDB on the system if the team has not yet done so. Download and install MongoDB Community Server here (

After installation, start the MongoDB server by running the MongoDB command. This will start the MongoDB daemon process, allowing connections to the database.

Choose a suitable MongoDB client or driver based on the programming language and environment. MongoDB provides official drivers for programming languages like Python, Java, and Node.js. Client applications such as MongoDB Compass are also available for GUI-based interactions.

Establish a connection and authenticate. Once connected, select the MongoDB database to work with, or if it doesn’t exist, MongoDB will create it when the user first writes data to it.

After successfully connecting to the MongoDB database, depending on requirements and permissions the user can perform various operations like inserting documents, querying data, updating documents, and deleting data.

How to Access a MongoDB Database—or Limit Access to MongoDB Databases

To access MongoDB databases, it is typically necessary to connect (see above discussion). Manage access to MongoDB databases with authentication and authorization mechanisms:
  • Enable authentication. MongoDB doesn’t require authentication by default, but it’s highly recommended that all users enable authentication to secure MongoDB deployments.
  • Create user accounts. These should detail specific roles and privileges.
  • Authorization and roles. Roles define which actions users are authorized to perform on databases, collections, and other MongoDB resources. Define custom roles, or use the built-in MongoDB role-based access control (RBAC) system to manage user permissions.
  • Assign roles to users. After defining them, assign roles to user accounts to grant specific privileges.
  • Restrict access. Following the principle of least privilege, when assigning user roles, grant the minimum necessary permissions for the tasks.
  • Configure IP whitelisting. Restrict access by specifying the IP addresses or CIDR blocks that are allowed to connect to the MongoDB server in the MongoDB configuration file.
  • Enable audit logging. MongoDB Enterprise Edition audit logging capabilities allow tracking and logging of user activities and administrative actions and monitor and identify any suspicious or unauthorized activities.

What Type of Database is MongoDB?

It is a document-oriented database.

Is MongoDB a relational database, or is MongoDB a NoSQL database?

It is a type of NoSQL database. The difference between MongoDB vs relational databases is the way they store data. Unlike traditional relational databases, MongoDB databases don’t store data with a classic tabular structure and fixed schema.

Instead, they store data in flexible, schema-less documents, typically using formats like JSON or BSON. This allows for readier representation of complex data types—a factor making MongoDB vs traditional databases especially appealing for developers using big data or anyone in need of real-time retrieval.

Are there MongoDB relational database capabilities?

While they might be implemented differently due to their document-orientation, MongoDB databases can offer some capabilities that are similar to relational databases.
  • MongoDB supports indexing. This improves query performance by enabling efficient data retrieval, similar to how indexes work in relational databases.
  • Query language. MongoDB still allows for querying data using conditions, projections, sorting, and aggregation—querying capabilities similar to those found in relational databases.
  • Data integrity and transaction support. MongoDB now offers multi-document transactions, which allow for atomic operations across multiple documents to help ensure data consistency, similar to transactions in relational databases.
  • ACID compliance. MongoDB aims to provide ACID compliance for transactional operations (atomicity, consistency, isolation, durability), ensuring similar data integrity and reliability to relational databases.
  • Join-like operations. MongoDB promotes denormalization and embeds data within documents to improve performance. However, it also offers capabilities for simulating join-like operations using aggregation framework, allowing users to combine data from multiple collections.

Additional MongoDB database use cases

Is MongoDB a NewSQL database?

No, but it is well-suited to some of the same applications. For example, the rapid read and write operations of the MongoDB NoSQL database are well-suited for the responsive, real-time user performance needed for online transaction processing (OLTP).

Is MongoDB a columnar database?

No. A wide column store or columnar NoSQL database organizes related data in column families rather than traditional rows, storing large amounts of data across a distributed column architecture.

Is MongoDB a graph database?

No. Although it has some graphing capabilities, there is no formal MongoDB graph database. However, it does offer some features and integrations that can be used to work with graph data.

  • Aggregation framework. This allows users to perform complex data transformations, aggregations, and computations within the database. While not specifically designed for graph processing, it can perform some graph-like operations and filter data based on graph-like patterns.
  • The $graphLookup aggregation stage. This enables graph-like queries such as finding connected data or finding paths between nodes in a collection.
  • Third-party integrations. Third-party tools and libraries such as ArangoDB provide graph database capabilities on top of MongoDB.
  • GraphQL. This query language for APIs allows clients to request only the data they need. GraphQL is often used with MongoDB to build APIs that expose graph-like data structures. This can provide a more flexible and efficient way to query and manipulate connected data.

Is MongoDB a shard database?
Yes. Sharding and MongoDB database replication ensure fault tolerance. And distributing data across multiple servers guarantees redundancy and allows MongoDB to handle large datasets with high throughput.

How to Set Up MongoDB Databases

After first connecting to MongoDB, the focus shifts to the data and operations:
  • Create databases and collections. Databases are containers for collections, which contain documents.
  • Insert data into databases and collections and perform operations. Start inserting data into your MongoDB databases and collections. Perform operations such as querying data, updating documents, and deleting data.
  • Backup and maintenance. Develop backup strategies and perform regular maintenance to ensure the integrity and availability of MongoDB databases.

How to Structure MongoDB Databases

MongoDB database architecture is flexible and schema-less. While each document in a collection can have a different structure, a schema design should reflect the application’s data model and query patterns.

Denormalizing data and embedding related information within a single document may optimize query performance, minimizing the need for complex joins and facilitating efficient data retrieval.

MongoDB supports compound, geospatial, multi-key, and single-field indexes. Create indexes on frequently queried fields to speed data retrieval and improve query performance.

MongoDB supports data sharding across multiple servers to load balance, improve scalability, and handle large datasets with high throughput. Plan sharding strategy based on application data distribution and access patterns.

MongoDB replica sets consist of primary and secondary nodes that asynchronously replicate data. If the primary node fails, MongoDB uses an internal election process to elect a new primary node. Configure replica sets to ensure data redundancy and fault tolerance.

Secure MongoDB deployments and architecture by enabling authentication and access control. Set appropriate roles and permissions to restrict access to databases and collections. Implement network security measures such as IP whitelisting and encryption to protect data in transit and at rest.

MongoDB adds more servers to scale horizontally, handling increased workload and data volume.

MongoDB database schema design points

The MongoDB database schema is naturally suited to accommodate diverse data types, and support changing requirements and evolving use cases over time.

The key features of the database design for MongoDB that allow for this flexibility include:

  • The document-oriented model that allows users to store data together as they feel it is related;
  • The lack of a rigid structure;
  • The ability to embed related data for more complex structures;
  • Effective versioning and migration strategies;
  • Extensible APIs; and
  • The ability to include customizable metadata.

How to Create a Database in MongoDB

To create databases with MongoDB, use the right platform to achieve specific goals. Here are some specific MongoDB create database approaches:

How to Create Databases in MongoDB with Drivers and Client Libraries

  • Connect to the server using the appropriate MongoDB supporting driver or client library based on the programming language (pymongo for Python, MongoDB Node.js Driver for Node.js)
  • Authenticate with valid credentials where enabled
  • Use the driver’s API to create a new database
  • Most MongoDB drivers offer interactive functionality for use with databases and collections

Can Users Create Database Commands in MongoDB Atlas?

MongoDB Atlas provides its own tools and interfaces for creating, querying, and managing databases, collections, and documents:

  • MongoDB Shell access. This offers an interactive JavaScript interface accessible from local machines or directly within the MongoDB Atlas web interface.
  • MongoDB Compass. This graphical user interface (GUI) for MongoDB provides a visual interactive capability between users, databases, and collections. Atlas and MongoDB Compass integrate seamlessly allowing users to connect to clusters and perform database operations via a user-friendly interface.
  • MongoDB drivers and client libraries. MongoDB Atlas is fully compatible with the drivers and client libraries for various programming languages needed to build applications that interact with Atlas clusters programmatically such as Python, Node.js, and Java. Install these drivers using package managers like npm (Node.js) or pip (Python) and connect them using the connection string in the Atlas web interface.
  • Atlas Data Explorer. This built-in tool for browsing and querying databases and collections directly within the Atlas web interface allows users to execute queries using MongoDB’s query language, and view query results.

Create MongoDB databases using MongoDB Shell

  • Connect to the MongoDB server using the mongo shell or a client application.
  • If authentication is enabled, authenticate with valid credentials.
  • To create a new database, simply switch to the desired database using the use command followed by the database name. If the specified database doesn’t exist, MongoDB will create it when you first write data to it.

Set up MongoDB databases using MongoDB Compass

  • Open MongoDB Compass and connect to the server.
  • Authenticate with valid credentials if required.
  • In the left-hand navigation pane, click on the “Databases” tab.
  • Click “Create Database” and name the new database.

How to Manage MongoDB Databases

There are several key, interrelated aspects of MongoDB local database management to be aware of across several areas:

How to check MongoDB databases for anomalies

This typically involves monitoring various metrics and performing periodic audits to identify irregularities and anomalies:

Use built-in monitoring tools or third-party monitoring solutions. Set alerts for abnormal values or thresholds to proactively detect anomalies and potential performance issues. Track key monitoring metrics:

  • database operations (index usage, query performance, and read/write throughput)
  • replication status (election outcomes, lag, and member health)
  • server status (CPU/memory usage, disk I/O, and network traffic)
  • storage utilization (data size, storage engine metrics)

Enable query profiling. Capture information about query execution times, operations, and index usage. Analyze query profiler output to identify slow-running and frequent queries, as well as those consuming excessive resources. Optimize queries with appropriate indexes, adjust configuration parameters based on profiling data, and rewrite queries as needed.

Enable auditing and logging. Log operations such as authentication events, application-level events, database commands, and administrative actions. Review audit logs regularly to detect unauthorized access, suspicious activities, or deviations from expected behavior.

Perform data validation checks. Use MongoDB’s schema validation feature to enforce data integrity and consistency at the database level. Implement application-level validation logic to verify data correctness, completeness, and adherence to business rules.

Conduct routine security scans and vulnerability assessments. Use scanning tools and security services to assess configuration settings, encryption protocols, authentication, and access controls to identify potential weaknesses.

Conduct periodic reviews and audits. Evaluate the MongoDB deployment in terms of its overall health, performance, architecture, access controls, configuration settings, and compliance with best practices. Review backup strategies and disaster recovery plans. Identify areas for improvement and implement corrective actions.

How to Clean MongoDB Databases

Cleaning MongoDB databases typically involves identifying and resolving the root cause of issues and performing corrective actions that restore database health and integrity:

Identify anomalies. Review monitoring logs and audit trails for deviations from expected behavior. Investigate slow query performance, resource bottlenecks, data corruption, or security breaches.

Perform data cleanup. Clean up unnecessary data that may add to performance issues or storage inefficiencies such as expired or redundant documents, collections, or indexes. Compact collections, archive historical data, or reclaim unused disk space to optimize data storage.

Analyze and optimize slow-running queries identified during monitoring and profiling. Improve query performance by adding appropriate indexes, rewriting queries, or restructuring data models. Review and optimize index usage to improve query execution efficiency and reduce resource consumption.

Resolve replication/sync issues if anomalies are detected in sharded clusters. Investigate replication lag, member synchronization issues, and consistency. Take corrective actions as needed, for example repairing data inconsistencies or reconfiguring replica sets.

Address security breaches identified during security scans or audits. Patch or update MongoDB server software to the latest version to mitigate known security vulnerabilities. Review and strengthen access controls, authentication mechanisms, encryption protocols, and network security settings.

Document and review the cleaning process. This should include actions taken, root causes identified, and lessons learned. Review and analyze the effectiveness of corrective actions and preventive measures to refine and improve them based on feedback and insights gained from cleaning activities.

How to Backup MongoDB Databases

There are several basic steps to backup MongoDB databases

Choose a method first. MongoDB supports several backup methods, including mongodump, filesystem or storage-level snapshots, and MongoDB Atlas’s built-in backup service.

Mongodump is a command-line utility for creating binary backups of MongoDB databases and collections in BSON format. Filesystem or storage-level snapshots create point-in-time backups of MongoDB data files.

MongoDB Atlas backup built-in capability provides automated backups, point-in-time recovery, and cloud-based storage. Define a backup schedule based on data retention policies, recovery point objectives, and business requirements. Then determine backup frequency—hourly or daily, for example—and how long backup data will be retained (from days to indefinitely).

Once the schedule is in place, execute backup operations using the chosen backup method. Encrypt and store backup data securely to ensure it doesn’t become corrupted, and to prevent unauthorized access and data loss.

After backups take place, maintain some best practices to ensure a robust backup strategy for MongoDB databases that ensures data protection, availability, and compliance with regulatory requirements:

  • Regularly test backup and restore procedures
  • Regularly validate the integrity and reliability of backup data
  • Perform recovery drills to ensure backups can be restored successfully and efficiently in the event of data loss or disaster
  • Monitor backup operations to track progress, completion status, and errors, alerts and notifications of failures, delays, or anomalies that require attention
  • Document backup procedures, including schedules, methods, storage locations, and recovery processes. Maintain documentation that is up-to-date and accessible to relevant stakeholders involved in backup and recovery operations.

How to Restore MongoDB Databases

This involves recovering data and restoring it to a MongoDB instance. The specific steps for restoring a MongoDB database typically vary:

  • Identify the backup source. This could be a backup file created using mongodump, filesystem or storage-level snapshots, or MongoDB Atlas.
  • Prepare for the restore process. Ensure the necessary permissions and access to the required backup files or snapshots are available. Verify that the target MongoDB instance has adequate storage space and resources to accommodate the restored data.
  • Select and execute the appropriate restore method based on the backup source.Use mongorestore to restore data from a backup created with mongodump; filesystem or storage vendor tools to restore data when using filesystem snapshots; and MongoDB Atlas Restore via the Atlas web interface to initiate a restore operation from the backup snapshot.
  • Verify restored data to ensure its integrity and completeness.Perform spot checks or query data to confirm that the restored database is consistent with the original state before the backup.
  • Update configuration settings as needed.It may be necessary to update authentication credentials or access controls to align with the restored database environment. Ensure that any dependencies or external integrations are properly configured to interact with the restored MongoDB database
  • Monitor and test the restored MongoDB database for anomalies. Conduct tests and validation checks to ensure that the restored database functions correctly and meets the desired performance and reliability criteria.

How to Delete MongoDB Databases

In MongoDB, delete databases using either the MongoDB Shell or a MongoDB management tool like MongoDB Compass.

Using MongoDB Shell:

  • Connect to the MongoDB instance using the MongoDB Shell (mongo command) or a client application.
  • Ensure the necessary permissions exist to delete databases—typically, the dropDatabase action must be present on the target database.
  • To delete the database, use “dropDatabase()” and the name of the database.

Using MongoDB Compass:

  • Open Compass and connect to the MongoDB instance.
  • Click on “Databases”.
  • Right-click on the database to delete and select “Drop Database”.
  • In response to the Compass prompt, confirm the irreversible deletion from the instance.

Delete databases with caution, and ensure any necessary backups are in place.

How to export MongoDB databases

In MongoDB export databases using the mongodump tool or Compass.

mongodump creates a binary export of the contents of a database into a specified directory. Open a terminal or command prompt using mongodump. Each database will be exported into its own subdirectory within the output directory.

Alternatively, using Compass to connect to the MongoDB instance, click on “Backups” in the left-hand navigation pane. Click on “Export Data” and select the databases and collections to export. Specify export options, such as the export format (JSON or CSV) and the destination directory. Start the export process and Compass will export the selected databases and collections into the specified directory in the chosen format.

Benefits of MongoDB Database

The key benefits of MongoDB include:

  • A flexible, document-oriented data model. MongoDB’s dynamic schema designs make it easy to store and manage diverse data types and structures.
  • Ease of development. Expressive APIs, rich queries, and a robust ecosystem of drivers, libraries, and tools simplifies application development and integration.
  • Agile development and rapid iteration. Schema-less architecture and flexible data models support agile development and rapid iteration of applications.
  • Automatic failover and high availability. Replica sets provide data redundancy, ensuring continuous availability and resilience against server failures.
  • Global distribution and data locality. MongoDB Atlas offers multi-region deployment options for improved performance and compliance.
  • Community and support. The open source MongoDB database nurtures an active community of users, contributors, and partners who generate resources, documentation, tutorials, and support to other users who want to succeed with MongoDB.

Does ScyllaDB Cloud Provide an Alternative to MongoDB Atlas?

Yes. In fact, in at least one independent benchmarking study, ScyllaDB Cloud outperforms MongoDB Atlas in 132 of 133 performance measurements.

For example, ScyllaDB Cloud (NoSQL DBaaS) provided up to 20 times higher throughput results compared to MongoDB Atlas for all applied workloads. ScyllaDB also achieved P99 latencies < 10 ms for insert, read and write operations for almost all scenarios.

In contrast, MongoDB Atlas achieved P99 latencies < 10 ms only for certain read operations, with MongoDB Atlas insert and update latencies up to 68 times higher than those achieved by ScyllaDB Cloud.

ScyllaDB Cloud achieved near linear scalability, while MongoDB Atlas showed less efficient horizontal scalability. Furthermore, ScyllaDB Cloud had a strong cost advantage with up to 19 times better price-performance ratio depending on the workload and data set size.

In total, ScyllaDB provides a great solution for applications that operate on terabytes of data that require high throughput (over 50K OPS) and predictable low latency for read and write operations.

Learn more about how ScyllaDB and MongoDB compare here.