Data is collected from multiple data sources across the enterprise and the edge using Amazon Kinesis Data Streams’ many SDKs with support for languages like Java, .NET, C++, python, Javascript, and others. Data persists and is sent to Amazon Simple Storage Service (Amazon S3) by Amazon Kinesis Data Firehouse. AWS Lambda can be used to enrich data prior to storage in Amazon S3. Initial data preparation and aggregation is performed using Amazon Athena. Prepared and aggregated data is stored in Amazon S3. Amazon SageMaker is used to train a forecasting model and create predictions of future behavior. These can be predictions for either statistical descriptions (for example sample counts and standard deviations) or business-oriented aggregat

http://chrome-extension://efaidnbmnnnibpcajpcglclefindmkaj/https://d1.awsstatic.com/architecture-diagrams/ArchitectureDiagrams/monitoring-streaming-data-with-mlv3-ra.pdf?did=wp_card&trk=wp_card