Question

Difficulty: EasyHigh-Performing Data Ingestion and Transformation Solutions

A retail company wants to design a serverless ingestion pipeline to collect real-time JSON clickstream data from its web application at a rate of 500 KB/s500\text{ KB/s}. The data must be converted into Apache Parquet format and saved in Amazon S3 for subsequent analysis. The solution should require minimal administration and operational overhead.

Which two AWS services or features should be used together to meet these requirements? (Select two.)

  1. Amazon Kinesis Data Firehose to ingest the clickstream data and deliver it directly to Amazon S3.Answer
  2. AWS Glue integration within Amazon Kinesis Data Firehose to perform format conversion from JSON to Apache Parquet.Answer
  3. C
    Amazon SQS standard queues to ingest and buffer the raw clickstream data while maintaining strict chronological message ordering.
  4. D
    Amazon Kinesis Data Streams configured with a single provisioned shard to automatically handle throughput peaks up to 10 MB/s10\text{ MB/s}.
  5. E
    AWS Lambda functions running continuously on a 24/7 schedule to poll a web server and handle the stream ingestion.

Answer

To ingest and transform JSON clickstream data with minimal operational overhead, use Amazon Kinesis Data Firehose to receive and write the stream directly to Amazon S3, and configure AWS Glue integration within the Firehose stream to handle the inline transformation to Apache Parquet.
Using Amazon Kinesis Data Firehose simplifies the ingestion pipeline by automatically writing stream data to Amazon S3. Its native integration with AWS Glue allows for serverless, zero-code format conversion from JSON to Apache Parquet before S3 storage, keeping operational overhead to a minimum.

Step-by-Step Solution

1
Identify the service that can ingest real-time streaming data and deliver it directly to Amazon S3 with minimal administration.
Amazon Kinesis Data Firehose is selected as it is a fully managed, serverless delivery stream.
It removes the need to manage custom consumer applications or polling infrastructure.
2
Determine the transformation mechanism for converting JSON to Apache Parquet within the ingestion pipeline.
Configure Kinesis Data Firehose to use the AWS Glue Data Catalog schema to convert JSON to Parquet inline before writing to S3.
This provides a serverless, no-code transformation mechanism that aligns with the requirement of minimal operational overhead.

Key Concept

Serverless data ingestion and inline transformation using Kinesis Data Firehose and AWS Glue.
Rate this question