MongoDB Atlas Data Federation - what data is being saved outside of the original data sources

Hi
I am looking at using Atlas Data federation to be able to query data from multiple databases. All of the data sources are clusters/DBs in the Atlas . Is any of the queried data being saved in AWS by the federation component. Looking to see if this works from out data privacy and localization point of view. All of our clusters are being hosted in Azure right now. Is there any caching of the data happens in the federated instance ? Where will the indexes be stored?
Thank you

Hello @Ilana_shapiro ,

Thanks for the question!

In short, no, there is no data saved in AWS by the federation service. Currently the infrastructure for federation is hosted in AWS, but all data is ephemeral in the system and is deleted once a query has finished.

There is no data that will be persisted in the federated instance.

But I do have some good news, native Azure support is coming soon too. Once we release it, it should be a very simple switch that allows you to switch to having all processing happen in Azure. The statement from earlier will still stand that is data is immediately released once a query completes, but there will be a slight benefit in terms of less Data Transfer costs.

There is no caching of customer data happening. For the use case of using data federation to query a cluster the indexes are stored on the cluster, they get used locally by the cluster when the cluster is returning results to the federated layer but they are not moved or copied.

Best,

2 Likes

This topic was automatically closed 5 days after the last reply. New replies are no longer allowed.