Introduction to MongoDB Architecture

mongod Process

The mongod process is the core database server responsible for data storage, query processing, and managing database operations.

• Handles CRUD operations on collections.
• Manages data files and indexes.
• Supports replication and sharding.

mongos Router

In a sharded MongoDB cluster, mongos acts as a query router, directing client requests to the correct shard based on the cluster metadata.

• Transparent to applications.
• Enables horizontal scaling by distributing queries.
• Coordinates query results from multiple shards.

Data Storage and Replication

MongoDB stores data in flexible, JSON-like documents using BSON format, allowing dynamic schemas.

To ensure high availability, MongoDB uses replica sets which replicate data across multiple servers.

• Documents are grouped into collections, similar to tables in relational databases.
• Replica sets consist of primary and secondary nodes for automatic failover.
• Writes go to the primary node and replicate asynchronously to secondaries.

Sharding for Horizontal Scaling

Sharding allows MongoDB to scale horizontally by partitioning data across multiple servers or shards.

Each shard holds a subset of the data, enabling the database to handle large datasets and high throughput.

• Shards are deployed as replica sets for fault tolerance.
• mongos routers direct queries to the appropriate shards based on shard keys.
• Config servers maintain metadata about the cluster's data distribution.