Exam AWS Certified Solutions Architect - Associate SAA-C03 topic 1 question 737 discussion

To keep replication in SYNC across all three regions, we use Bi-directional. Multi-Region Access Point for video streaming and uploads. -> uploads to nearest Low latency region and Bi-directional replication will keep other two regions in SYNC this reducing the upload and streaming latency

Question does not make a lot of sense with this responses 1) If the goal is to modify the app as little as possible, options D and E don't present much of a difference. If you modify the app for uploads, it will cost you the same to modify it for streaming (even more so with an MRAP, since you only need to change the global upload endpoint, which is a trivial change). 2) It is not suggested that the need is "to upload from anywhere in the world" but rather to copy to other regions for streaming (download), which an MRAP does not do per-se. 3) If the idea of the question is to create an MRAP (D or E) and then set up replication rules in the MRAP (C), it doesn’t make sense to replicate from US-West or from Asia-Pacific, but only the other way around. 4) Option A stands on its own (but it’s a multiple-choice answer), and even though the initial replication latency is slow, uploading video tutorials is not critical. 5) Option B involves unnecessary latency.

The data is stored in an Amazon S3 bucket in the us-east-2 Region implies the content developer is in us-east-2 region. If the content developers are in the us-east-2 Region, then the solution is A and D.

This is an example of the kind of joke problem AWS provides. The information needed to derive justification is vague and it is not clear which region the uploading developer will be using. Time spent on these problems is wasted.

I agree with @kbgsgsgs and here's why: What the scenario is about: Currenly, videos are in us, so they're uploaded to us. Then, we want to replicate them to eu and ap. People near eu and ap should be fast delivered with the videos they need. So with "FEWEST changes", all we need is replicate from us to both eu and ap, which is option B. Option C works but not fewest changes. The uploads are directly to us, so option E actually induces more latency. You may argue that if we can upload videos directly to eu/ap, it would be faster. Yes, but it requires modifying the application for both streaming and upload workflows, so much more complicated than option D, and definitely not fewest changes.

My personal take: C & D C => Although stem did not specify which region S# bucket the developers will upload their contents, this will allow new contents in 1 S3 bucket to replicate over to the other 2 S3 buckets. D => uploads needs to be performed to the first region only (Closest???) and accessed by remaining two S3 buckets.

They are simply trying to replicate to a new S3 bucket. I don't see why it needs to be bidirectional. Also, since the problem assumes that the content developer with permission is the one uploading, it seems like there needs to be a way to centralize the upload without modifying the application.

Since developer upload video to us-east-2, by configuring one-way replication directly from us-east-2 to eu-west-2 and from us-east-2 to ap-southeast-1, you ensure that each region has the latest data without additional replication hops.

AnswerCE From my understanding Video uploads can happen near new regions hence to speed up that operation we need to upload to nearest region, hence I would choose option E, and for the same reason we need to be able to replicate data from any of new region to old one and oposite, hence we we need bidirectional (two-way) replication

FEWEST changes to the application D -> MRAP can upload the appropriate S3 bucket C -> two-way -> to worry about anything obs: I believe this question dubious, amphibological

There is no information where the upload should be performed. If files will be uploaded to first region then: AD because: A -> content uploaded to the primary bucket in us-east-2 is automatically replicated to the other regions, minimizing latency for users accessing content near those regions. D -> uploads needs to be performed to the first region only and accessed to remaining two Otherwise CE

Correct answer CE