Which feature allows MongoDB to scale effectively?

The ability to scale effectively is primarily attributed to sharding in MongoDB. Sharding is a method of distributing data across multiple servers or shards, which allows the database to handle larger datasets and more write and read operations by balancing the load among various nodes. This horizontal scaling approach enables MongoDB to manage significant volumes of data and high throughput without compromising performance.

In contrast, while features like the Aggregation Framework provide powerful data manipulation and querying capabilities, they do not directly contribute to the scaling of the database infrastructure itself. Document DB locking pertains to concurrency control and is important for data integrity but does not influence how data is distributed or accessed across multiple servers. Cross-region replication enhances data availability and resilience but is more about data redundancy and failover rather than scaling out operations effectively.

Thus, sharding stands out as the key feature that empowers MongoDB to scale efficiently by distributing data and workload seamlessly across multiple instances.