Atlas Search runs a new process, called
mongod process on each host in your
mongot maintains all
Atlas Search indexes on collections in your Atlas databases.
The amount of CPU, memory, and disk resources
depends on several factors, including our index configuration and
the complexity of your queries.
Atlas Search alerts measure the amount of
CPU and memory used by Atlas Search processes.
The following pages contain recommendations that you can apply when configuring your Atlas Search index and defining your Atlas Search queries to improve performance:
To fix and monitor Atlas Search issues, see the following pages:
Upgrading to NVMe Storage
Upgrading an Atlas cluster to use NVMe triggers an
initial sync on that cluster.
Atlas Search is unavailable until both the
finish their initial syncs.