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