Dynamic Data Masking on S3 data with Redshift Spectrum

Question

Is it currently possible to perform dynamic data masking on data stored in Amazon S3 buckets when querying using Redshift Spectrum?

I understand, based on the documentation-
https://docs.aws.amazon.com/redshift/latest/dg/t_ddm.html, that masking policies cannot be created on views, external tables, and materialized views at this time. 
However, I am interested in achieving dynamic data masking on s3 data at query time using policies defined within Redshift with Redshift Spectrum.

Are there any alternative approaches to achieve a similar level of data protection for S3 data accessed through Redshift Spectrum?

Also,
In Redshift, dynamic data masking policies are typically associated with users and roles created within the Redshift editor. Can we leverage real-world users and roles (e.g., users authenticated via AWS IAM) and attach/create masking policies accordingly?
If yes, how to do it?

Answer

If you are looking to mask data when retrieved, you can try Object Lambda: https://aws.amazon.com/blogs/machine-learning/protect-pii-using-amazon-s3-object-lambda-to-process-and-modify-data-during-retrieval/

Dynamic Data Masking on S3 data with Redshift Spectrum

Relevant content