Hi team, first post, let me know if it provides a good explanation.
I'd like to know a way to minimize the effort for data ingestion.
We have two options as follows:
(1) csv files from a file server located outside our organization (in the supplier proprietary cloud)
(2) no-sql MongoDb inside our organization
Considering that we are transitioning to AWS and we have available S3, Lambda, Glue, Athena, Redshift (not yet defined all the services).
Considering that in our organization we have a team that manages infrastructure, network, security for us and we need to build data ingestion, transforming and load (ELT/ETL).
Could you provide pros and cons about each of above options?
Also, provide an example of architecture for further studies?