Are you running DMS loading into the raw layer? Is the destination database part of the current project? Are you storing the data?

AWS Database Migration Service (DMS) is used to replicate data from a source database to a target database with minimal downtime. In a typical data lake architecture, the raw layer is the first landing zone where data is ingested in its native format. When you run a DMS task to load data into the raw layer, you configure the source endpoint (e.g., an on‑premises Oracle database) and the target endpoint (e.g., an Amazon S3 bucket or an Amazon Redshift cluster). The destination database or storage must be part of the current project’s data architecture so that downstream pipelines can discover and process it. Storing the data means persisting it in a durable, cost‑effective storage service (S3, Redshift, or a data warehouse) and registering it in the Glue Data Catalog if you plan to run ETL jobs. Key considerations include choosing the right replication mode (full load, ongoing changes, or both), setting up proper IAM roles, and ensuring network connectivity (VPC endpoints or VPN). Properly configuring DMS ensures data integrity, minimal latency, and a clean separation between raw and curated layers.