Answer: B, C, E
Option A is incorrect because the data extraction agents should NOT be connected to the production Cassandra cluster. The extraction process performed by the data extraction agents imposes an additional overhead load on the Cassandra database. To avoid a negative impact on the operation and performance of the source production cluster, a clone datacenter containing a copy of the production data should be created. SCT should be configured to connect to this clone cluster.
Option B is CORRECT because SCT data extraction agents are required when the source and target databases are significantly different and require complex data transformations. This is the case for migrating data from Cassandra to DynamoDB. Data Extraction Agents come as part of the SCT installation.
Option C is CORRECT because the extraction process performed by the data extraction agents imposes an additional overhead load on the Cassandra database. To avoid a negative impact on the operation and performance of the source production cluster, a clone datacenter containing a copy of the production data should be created. SCT should be configured to connect to this clone cluster.
Option D is incorrect because Kerberos is not required for connecting and authenticating to a Cassandra cluster using SCT.
Option E is CORRECT because the S3 bucket is required to store the output of the data extraction agent conversion and migration process.
Option F is incorrect because SCT uses AWS access key and secret key for AWS authentication.
Reference:
https://docs.aws.amazon.com/SchemaConversionTool/latest/userguide/agents.cassandra.html