Exam AWS Certified Solutions Architect - Associate SAA-C03 topic 1 question 888 discussion

B - C are out = Glue doesn't support real-time data processing. D - Why would you use Kinesis data ANALYTICS to ingest clickstream data instead of Amazon Kinesis DATA STREAM?

This one is very tricky, need to read the context carefully: "The company needs a scalable solution that can adapt to varying levels of traffic" -> Both Firehouse and Stream are scalable. But, Firehose is automatic where Stream is not. However, the question does NOT say it should be automatic and Glue is not support real-time analysis. Thats why go for A. B is very close to.

AnswerA Apache Flink (previously known as Amazon Kinesis Data Analytics) seems to not allowing sent data directly to Lambda... Glue is allowing to integrate data from couple of sources in to one. Hence I think A is correct answer https://docs.aws.amazon.com/lambda/latest/dg/with-kinesis.html https://aws.amazon.com/kinesis/data-streams/features/?nc=sn&loc=2

Seem AWS Glue does not support process data in real time. I vote for A

Both Kinesis Data Streams and Firehose are scalable but Firehose offers automated scaling. I vote fore B

I think Apache Flink (previously known as Amazon Kinesis Data Analytics) would also be fine, but as here it wants to combine it with Lambda, I would rather opt for Kinesis Data Streams + Lambda, so A, because of the figure on this page: https://aws.amazon.com/kinesis/