Exam AWS Certified Data Analytics - Specialty topic 1 question 151 discussion

Near/Real-time analytics on live data is always done via KDA. Once the data is stored it cannot be seen as live data anymore. Arguments regarding operational overhead are relevant and will also mix things up for me, but I think here we need KDA.

I am not clear about C or D. For C: it use API gateway as a proxy. ALthough it can, as mentioned beow by "Teraxs". But Is it a good idea to use API gateway for streaming data? For it's pricing, it will cost about for every million message. If the company has millions of customers, uploading activity stream every second. The cost will be huge. For D: The question says they need analyze live data. So it needs kinesis analystics. But D doesn't mention about that. And answer D use kinesis firehose, it is also not near real-time. But, anyway, C will meet the requirement, but D cannot. So I choose C.

FSDFASDF

D is not correct. D cant solve realtime anallitics.

I vote for D. D has the least operational overhead and fulfill the purpose. C need to manage KDS shard, API gateway, etc.

"the processed data must be stored", so it is C. D stores the raw data directly to S3

Stored PROCESSED data, not raw data, hence C instead of D!

D: I passed the test

I will go with C. Though D has least operational overhead, Athena can't be considered as the solution for Analytics

D sounds right

D has the least operational overhead

I vote for C. Because the company needs to perform analytics on LIVE data, I don't think D can be the answer. D analyzes the data after it is stored in S3 and does not perform analytics on live data.

C because it clearly says the company wants to ingest and perform analytics near-real time then the processed that is stored. So its KDA for near-real time analtics, although KDS is real-time, but it can also deliver near-real time.

The key to selecting C instead of D is: "the processed data must be stored…". In C, the data has been processed through KDA, then saved in S3. But in D, data is from the application to S3 directly, not being processed yet, Athena comes later. So I go with C.

for the purpose of the exam, near-real time is generally KDF. Have to agree that arguments can be made for C

Option D

For near real-time and with least operational overhead D is better answer