Data Transfer
On this page
Sources of Data Transfer Costs
Atlas data transfer costs depend on the cloud service provider hosting your database deployment. Atlas tabulates data transfer costs daily.
Clusters
Multi-region clusters might have higher data transfer costs depending on the number and location of additional regions, as well as the number of clusters deployed to each region.
Atlas doesn't charge for data transfer for incoming or outgoing
data on M0, M2, or M5 cluster tiers.
Serverless Instances
Serverless instances incur costs for data transfer to and from the virtual machine responsible for backing up and restoring data.
Cloud Backup Snapshot Export
Atlas charges $.125 per GB of data exported to the AWS S3
bucket. Atlas compresses the data before exporting. To estimate the
amount of data being exported, add up the
dataSize
of each database in your cluster. This total should correspond to the
uncompressed size of your export, which would be the maximum cost
incurred.
To learn more about cloud backup snapshot export, see Export Cloud Backup Snapshot.
Live Migrate Your Data to Atlas
MongoDB hosts and operates the free Atlas Live Migration Service to help users migrate existing MongoDB databases to MongoDB Atlas. MongoDB doesn't charge for any incoming data transfers to an Atlas database deployment. Learn more about migrating to Atlas.
How to Reduce Data Transfer Costs
The vast majority of Atlas customers spend less than 10% of their budget on data transfer. If you are spending significantly more, some of these optimizations may reduce your data transfer costs:
Check all applications and processes that access your data for inefficiencies. Ensure that queries do not:
- Re-read data that already exists on the client.
- Re-write existing data to your database deployment.
Ensure that queries originate from the same cloud region and provider as your database deployment whenever possible.
When cross-region queries are necessary:
- Ensure read queries use the preference "nearest."
- Source write queries from your Highest Priority Region whenever possible. For more information on region priorities, see Electable Nodes for High Availability.
- Use the Aggregation Framework to preprocess
your data before you transfer it. For example, you can project
document fields using the
$projectaggregation stage to reduce the size of a document before you transfer it. - Ensure that your client driver uses wire protocol compression to communicate with MongoDB. Atlas always compresses intra-cluster communication. To learn how to configure your driver, refer to your Driver's Documentation.
Queries from non-cloud environments into Atlas, across cloud providers, or between continents on the same cloud provider incur the greatest data transfer costs.