EFS Multi-AZ with cross-Region replication for a shared Linux application dataset

SAP-C02 By islever Editorial Last updated

Design a solution to ensure business continuity. Design a strategy to meet reliability requirements.

Short answer

A Deploy a Multi-AZ EFS file system with provisioned throughput and replicate it to a file system in the DR Region for the sub-hour RPO.

Question

A company is deploying a new web-based application and needs a storage solution for the Linux application servers. The company wants to create a single location for updates to application data for all instances. The active dataset will be up to 100 GB in size. A solutions architect has determined that peak operations will occur for 3 hours daily and will require a total of 225 MiBps of read throughput. The solutions architect must design a Multi-AZ solution that makes a copy of the data available in another AWS Region for disaster recovery (DR). The DR copy has an RPO of less than 1 hour. Which solution will meet these requirements?
  1. Deploy a new Amazon Elastic File System (Amazon EFS) Multi-AZ file system. Configure the file system for 75 MiBps of provisioned throughput. Implement replication to a file system in the DR Region. Correct Answer
  2. Deploy a new Amazon FSx for Lustre file system. Configure Bursting Throughput mode for the file system. Use AWS Backup to back up the file system to the DR Region.
  3. Deploy a General Purpose SSD (gp3) Amazon Elastic Block Store (Amazon EBS) volume with 225 MiBps of throughput. Enable Multi-Attach for the EBS volume. Use AWS Elastic Disaster Recovery to replicate the EBS volume to the DR Region.
  4. Deploy an Amazon FSx for OpenZFS file system in both the production Region and the DR Region. Create an AWS DataSync scheduled task to replicate the data from the production file system to the DR file system every 10 minutes.

Key Takeaways

Amazon EFS is the only shared POSIX file system in the options, supports Multi-AZ within a Region, and can replicate to a file system in a second Region, which together satisfy both the shared-mount and the DR requirements.

Linux application servers need a single shared location for application data, sized at up to 100 GB, sustaining 225 MiBps of reads during a three-hour daily peak, replicated to another Region with an RPO under one hour.

Common mistake: Assuming a DataSync scheduled task can run every 10 minutes. DataSync schedules have a minimum interval of one hour, so option D cannot meet an RPO of less than one hour regardless of how well the rest of the option reads.

Expert Analysis

Why the Answer Is Correct

Amazon EFS gives every EC2 instance a shared NFS mount that scales automatically, and a Multi-AZ file system spreads access across Availability Zones for high availability. EFS replication to a file system in the DR Region keeps a continuously updated copy, and provisioned throughput of 75 MiBps bursts well past the 225 MiBps peak for the three-hour window. EFS is also Linux-native NFS, which the CMS requirement demands.

Why the Other Options Are Wrong

B: Bursting Throughput is an EFS feature, not an FSx for Lustre feature, and AWS Backup is not a continuous replication mechanism, so the RPO fails. C: EBS Multi-Attach works only within an Availability Zone and cannot be replicated across Regions, and the gp3 volume is an attachable block device, not a shared file system. D: DataSync task schedules support a minimum interval of one hour, so a ten-minute schedule is not possible.

Community Comment Notes

The community split 68 to 32 for A against D, and the deciding factor was the DataSync one-hour minimum interval cited from the AWS task scheduling documentation, which rules out the ten-minute schedule in D.

Official Reference

https://docs.aws.amazon.com/efs/latest/ug/efs-replication.html

This link opens the official vendor documentation in a new tab.

Community Votes

A
68%
D
32%

68% of anonymous learners picked answer A. Votes are pick records left by other test-takers — they are not the verified answer.

Discussion 12 contributions

Compiled from public community threads and reader submissions, lightly edited for clarity. Handles and vote tallies are shown as recorded, and comments reflect the views of individual contributors.

From public community threads 12

  1. e4bc18e 👍 20
    So practically everyone here is wrong. because it is A. Here is why B is wrong because one there is no such thing as bursting mode for Lustre that is an EFS thing, but also Backup will not work for the RPO. C is wrong obviously because GP3 can't be shared. D is wrong because Datasync tasks cannot be scheduled for any more frequent then hourly so no D is wrong because you cannot schedule data sync tasks less then hourly so you don't meet the RPO. So all of those are easily wrong because they have bad information. They fooled everyone on A because all they say is the 'Active working set is 100GB" not the entire filesystem. EFS accumulates bursting credits so for every 100GB of filesystem size you can burst up to 300MiBps for up to 72 minutes. So you provision 75MiBps because that would average out over time so you aren't being overcharged for the provisioned size.
  2. pangchn 👍 4 Answered D
    D a sneaky question since my first impression is go for A but it is wrong due to the 75M throughput mode. What's the calculation here? one region has 3 AZ? so 75x3=225?. EFS is not provisioned in that way. Even that, the 225 is the total throughput where question asked 225 for read. Implied the total would be more like 225+XXX. Anyway, A is wrong. https://docs.aws.amazon.com/efs/latest/ug/performance.html C is wrong since EBS multi attach don't support gp3 https://docs.aws.amazon.com/ebs/latest/userguide/ebs-volumes-multi.html
  3. Helpnosense 👍 4 Answered A
    A. EFS support cross region replication. e4bc18e already point why D is wrong.
  4. VerRi 👍 3 Answered A
    D involves managing separate file systems that do not natively offer a "single location" experience across regions without additional configuration and replication mechanisms.
  5. vip2 👍 3 Answered A
    A scheduled task runs at a frequency that you specify, with a minimum interval of 1 hour. https://docs.aws.amazon.com/datasync/latest/userguide/task-scheduling.html
Show 7 more comments
  1. titi_r 👍 2 Answered D
    D is correct. "You can use DataSync to transfer files between two FSx for OpenZFS file systems, and also move data to a file system in a different AWS Region or AWS account. You can also use DataSync with FSx for OpenZFS file systems for other tasks. For example, you can perform one-time data migrations, periodically ingest data for distributed workloads, and schedule replication for data protection and recovery." https://docs.aws.amazon.com/fsx/latest/OpenZFSGuide/migrate-files-to-fsx-datasync.html
  2. CMMC 👍 1 Answered D
    Amazon FSx for OpenZFS is a fully managed file system service that supports native replication between regions, making it well-suited for DR scenarios with a low RPO requirement. Using AWS DataSync for replication every 10 minutes ensures that the DR copy stays up to date with minimal data loss. This solution provides the required read throughput, data replication, and DR capabilities with less operational overhead.
  3. adelynllllllllll 👍 1
    D: The throughput is related to size of the EFS, but the question said the active set of the data will be only up to 100GB, with that size, the throughout will be lower than requested. so D:
  4. ovladan 👍 1 Answered B
    https://docs.aws.amazon.com/fsx/latest/LustreGuide/performance.html#fsx-aggregate-perf
  5. Dgix 👍 1 Answered D
    D is the answer. A would also have worked.
  6. trungtd 👍 4 Answered A
    big thank to e4bc18e
  7. Zas1 👍 3 Answered A
    A Solution write by e4bc18e

Reader comments & corrections

No reader corrections yet — spotted an error or have a note? Share it below.

Log in to comment, report an error, or add a note about this question.

Submitted for moderation before publishing. Keep it helpful and respectful.

Modernize centralized firewall appliances with a Gateway Load Balancer and Auto Scaling

S-grade

Use Lambda over a CloudWatch Logs stream to count unique users per client

S-grade

Serve global game assets with Route 53 latency records and health checks

S-grade

Reach a cross-account S3 data lake through a gateway endpoint and an access point bucket policy

S-grade

Practice All SAP-C02 Questions

Access 85 questions with complete answers and detailed explanations.

View Full SAP-C02 Practice Test →

← Back to SAP-C02 Study Guide