Explanation
This is what it means to build a trust-based, first-party data asset. It involves respecting the privacy and preferences of individuals who share their data with an organization and ensuring that the data is used in a responsible and ethical way. References:https://www.salesforce.com/products/genie/overview/

These two phases are distinct phases of data model management in Data Cloud. Data ingestionis the process of bringing data from different sources into Data Cloud using connectors or APIs. Data modeling is the process of mapping the ingested data to the Data Cloud canonical model or creating custom data model objects.
References: https://help.salesforce.com/s/articleView?id=sf.c360_a_data_cloud_data_model.htm&type=5

The maximum limit of the number of records for Full Refresh Extract Method is 10 million. Full Refresh Extract Method is an option for extracting data from Marketing Cloud to Data Cloud. It extracts all records from a data extension regardless of when they were created or modified. If the number of records exceeds 10 million, an error occurs and no records are extracted. Reference: Extract Methods

This configuration will ensure thatData Cloud only ingests new files from the Amazon S3 bucket and updates or inserts new records into the data stream without deleting existing records.
References:https://help.salesforce.com/s/articleView?id=sf.c360_a_data_cloud_amazon_s3.htm&type=5

This is the first step to set up and configure a Data Cloud instance after it has been provisioned. The customer needs to connect to the Salesforce CRM org that Data Cloud is provisioned in, using the Salesforce CRM connector. This will allow Data Cloud to access the CRM data and map it to the Customer 360 data model. Reference: [Salesforce CRM Connector], [Data Cloud Setup]