The MongoDB Deployment Model

MongoDB can run on a variety of platforms, including Windows, macOS, and different flavors of Linux. You can install MongoDB on a single machine or a cluster of machines. Multiple machine installation provides high availability and scalability. The following list details each of these installation types:

Standalone

Standalone installation is a single-machine installation and is meant mainly for development or experimental purposes. You can refer to the Preface for the steps to install MongoDB on your system.

Replica Set

A replica set in MongoDB is a group of processes or servers that work together to provide data redundancy and high availability. Running MongoDB as a standalone process is not highly reliable because you may lose access to your data due to connectivity issues and disk failures. Using a replica set solves these problems as the data copies are stored on multiple servers. It requires at least three servers in a cluster. These servers are configured as the primary, secondaries, or arbiters. You will learn more about the replica set and its benefits in Chapter 9, Replication.

Sharded

Sharded deployments allow you to store the data in a distributed way. They are required for applications that manage massive data and expect high throughput. A shard contains a subset of the data, and each shard must use a replica set to provide redundancy of the data that it holds. Multiple shards working together provide a distributed and replicated dataset.