Exam AWS Certified Solutions Architect - Associate SAA-C03 topic 1 question 5 discussion

Concurrent or at the same time key word for EFS

Ebs doesnt support cross az only reside in one Az but Efs does, that why it's c

Implement Amazon EFS as shared storage for all EC2 instances behind the ALB. This ensures high availability, scalability, and consistent data access across Availability Zones with minimal operational overhead.

EFS is best suited for multi AZ and EC2 distribution

c is correct

Problem each EC2 instance has its own EBS volume, and EBS volume is specific for AZ specific so when user send request to specific ec2 instance they only see the response from each EBS volume not all because EBS volume do not sever for cross region problem C: use EFS because EFS a fully managed and network file system can be mounted to mutiple ec2 instances across mutiple AZ so user can see all documents response from EFS instead of subset of documents A. Copy the data so both EBS volumes contain all the documents is not correct because when user send request to one instance on a specific AZ they only see the documents from that EBS AZ not both

C is correct

The correct anszwer to this question is C question EBS volumes are AZ locked so when transferring the architecture to a different AZ the data can no longer be received by users whereas with EFS volume we can EC2 instances across AZs

My immediate understanding was the fact that EFS is multi az and will make it less complex as a solution for this.

Not A: because it requires sync between the EBS volumes which is complex and not scalable. Not B: It's not scalable if the LB only directs user to one instance Not D: Impractical. The application will need to have a mechanism to merge the responses coming from 2 instance

Use common storage like EFS so that the instances can access the document without any hassle and without being uploaded into them separately

Amazon S3 is the optimal solution for this scenario. It provides centralized, scalable, and highly available storage, ensuring that all users can see all their documents regardless of which instance they are routed to.

The answer is C. Copy the data from both EBS volumes to Amazon EFS. Modify the application to save new documents to Amazon EFS. The current architecture is using two separate EBS volumes, one for each EC2 instance. This means that each instance only has a subset of the documents. When a user refreshes the website, the Application Load Balancer will randomly direct them to one of the two instances. If the user's documents are not on the instance that they are directed to, they will not be able to see them.

To ensure that users see all of their documents at once, the solutions architect should propose Option C: Copy the data from both EBS volumes to Amazon EFS. Modify the application to save new documents to Amazon EFS. Option C involves copying the data from both EBS volumes to Amazon Elastic File System (EFS), and modifying the application to save new documents to EFS. Amazon EFS is a fully managed, scalable file storage service that allows you to store and access files from multiple EC2 instances concurrently. By moving the data to EFS and modifying the application to save new documents to EFS, the application will be able to access all of the documents from a single, centralized location, ensuring that users see all of their documents at once. Overall, Option C is the most effective solution for ensuring that users see all of their documents at once.

Keyword: second EC2 instance and EBS volume. They could see one subset of their documents or the other, but never all of the documents at the same time. EBS: attached to one instance (special EBS io1, io2 can attached to multiple instances but not much) EFS: can attached to multiple instances A: Incorrect - EBS volumes don't have function to copy data from running EBS volume to running EBS volume. B: Incorrect - We can use sticky session to forward same user to the same server but when user lose the session the user might be forward to another server. C: Correct - Because 2 instance now point to one EFS data storage, user will see both data. D: Incorrect - We only use Traffic Mirroring to sent request to both servers. Application Load Balancer don't support send request to both servers because it's design it balance workload between server. And also ALB cannot combine document from both servers and return.

Option A is not a good solution because copying data to both volumes would not ensure consistency of the data. Option B would require the Load Balancer to have knowledge of which documents are stored on which server, which would be difficult to maintain. Option C is a viable solution, but may require modifying the application to use Amazon EFS instead of EBS. Option D is a good solution because it would distribute the requests to both servers and return the correct document from the correct server. This can be achieved by configuring session stickiness on the Load Balancer so that each user's requests are directed to the same server for consistency. Therefore, the correct answer is D.

Ans C. Altho' A could do it, it would require a manual operation; the clue is "better scalability and availability" - EFS does that automatically