Home > Storage > PowerFlex > White Papers > MongoDB on Dell PowerFlex with NVMe over TCP > MongoDB architecture
MongoDB is an open-source NoSQL database that supports JSON-type data structures. MongoDB can be used to store a large volume of data and has a schema-less data model. It supports vertical and horizontal scaling. MongoDB uses sharding to support deployments with large datasets and high-throughput operations.
MongoDB supports multiple storage engines, as different engines perform better for specific workloads. WiredTiger is the default storage engine which is well suited for most workloads.
The following figure shows the high-level architecture of a MongoDB sharded cluster:
Figure 5. MongoDB sharded architecture
A MongoDB sharded cluster consists of the following components:
A typical MongoDB production deployment has secondary copies of each shard to increase the cluster resiliency. These copies are seen as a replica set with a default of three copies, which means one primary and two secondary copies. Replica Sets are the solution for a single point of failure. MongoDB uses a primary-secondary architecture. The inside mechanism of each shard has a replica set. For more information about MongoDB replica sets, see the MongoDB replica set documentation. Read-preference and write-concern parameters can be used to determine if these copies can be used to service reads.
For this whitepaper, secondary copies have not been configured in the MongoDB cluster. For more information, see MongoDB deployment.