AWS Machine Learning Engineer - Assocciate - Data ingestion

What are the ML supported data ingestion types?

Real-time and Batch

When is real-time ingestion used?

used for time sensitive data

when is batch ingestion used?

historical data that is not time sensitive and can be processed in batches

What services support real-time data ingestion for ML?

Kinesis, MSK, MSF and Data Firehose

What is Kinesis and when is it used?

Kinesis is a streaming data service that can ingest and process in real-time data streams from different sources. It is used when integrating with AWS services.

What is MSK and when is it used?

MSK is a managed streaming service for apache Kafka. It is used for kafka-centric environments.

What is MSF and when is it used?

Managed Service for Apache Flink is a streaming service that integrates with the apache Flink framework. It supports streaming to sagemaker for real-time inferences.

What is Data Firehose and when is it used?

Is a managed service that sends real-time streaming data to S3 or amazon redshift. It is used when you need to deliver streaming data to an output without any setup or management.