AWS Certified Solutions Architect – ProfessionalContinuously Improve Existing SolutionsHard

A financial institution processes millions of transactions daily. Their existing on-premises data warehouse struggles to ingest and process data fast enough to provide near real-time analytics for fraud detection. The current batch processing system has a latency of several hours, making it ineffective for identifying fraudulent activities as they occur. The company needs a highly scalable, fully managed data analytics platform on AWS that can ingest and process high-volume, streaming transaction data with sub-second latency, allowing for immediate fraud detection. Which AWS service combination should the Solutions Architect recommend to build this real-time data platform?

  1. AAmazon S3 for data lake, AWS Glue for ETL, and Amazon Athena for querying.
  2. BAmazon Kinesis Data Streams for ingestion, Amazon Kinesis Data Analytics for Apache Flink for processing, and Amazon Redshift for analytics.
  3. CAmazon MSK for ingestion, Amazon EMR for processing, and Amazon QuickSight for visualization.
  4. DAWS DataSync for data transfer, Amazon EC2 for processing, and Amazon DynamoDB for storage.
Show answer & explanation

Correct answer: B. Amazon Kinesis Data Streams for ingestion, Amazon Kinesis Data Analytics for Apache Flink for processing, and Amazon Redshift for analytics.

This combination provides a robust, fully managed real-time analytics pipeline. Kinesis Data Streams ingests high-volume streaming data. Kinesis Data Analytics for Apache Flink processes this data with sub-second latency for real-time fraud detection. Amazon Redshift can then be used for historical analysis and aggregated reporting.

Why the other options are wrong

  • A. This architecture is primarily for batch processing and would not meet the sub-second latency requirement for real-time fraud detection.
  • C. Amazon MSK (Managed Streaming for Apache Kafka) and Amazon EMR (Elastic MapReduce) can handle streaming data but typically involve more operational overhead than a fully managed Kinesis Data Analytics for Apache Flink solution for the real-time processing core. QuickSight is for visualization, not the core processing engine itself.
  • D. AWS DataSync is for large-scale data transfers, not streaming ingestion. EC2 requires significant management, and DynamoDB is a NoSQL database, not a data warehouse for complex analytics queries.

Real-time Streaming Analytics with Kinesis

A fully managed AWS solution for ingesting, processing, and analyzing high-volume streaming data with low latency to derive immediate insights.

  • Amazon Kinesis Data Streams for high-throughput data ingestion.
  • Amazon Kinesis Data Analytics for Apache Flink for real-time processing and analysis.
  • Suitable for immediate decision-making like fraud detection.

Memory trick: Kinesis Streams the data, Flink Analyzes it instantly, Redshift stores the results.

More Continuously Improve Existing Solutions questions