Which solution will meet these requirements MOST cost-effectively?
Create an Amazon S3 bucket that uses an Intelligent-Tiering lifecycle policy. Copy all files to the S3 bucket. Update the application to use Amazon S3 API to store and retrieve files.
Deploy Amazon FSx for Windows File Server file shares. Update the application to use CIFS protocol to store and retrieve files.
Deploy Amazon FSx for OpenZFS file system shares. Update the application to use the new mount point to store and retrieve files.
Create an Amazon S3 bucket that uses S3 Glacier Flexible Retrieval. Copy all files to the S3 bucket. Update the application to use Amazon S3 API to store and retrieve files as standard retrievals.
Explanations:
Using Amazon S3 with an Intelligent-Tiering lifecycle policy allows for automatic cost optimization by moving objects between different storage classes based on access patterns. This reduces costs compared to using EFS, especially for infrequent access files. The application can be updated to use the S3 API, which is feasible and cost-effective.
Deploying Amazon FSx for Windows File Server introduces higher costs due to licensing and operational overhead compared to Amazon S3. The CIFS protocol is also less efficient for cost optimization compared to using S3 for storage.
Amazon FSx for OpenZFS provides features for managing file systems but does not offer significant cost savings compared to Amazon EFS or S3. The transition to a new file system may also involve more operational complexity without substantial cost benefits.
While S3 Glacier Flexible Retrieval offers lower storage costs, the retrieval process is not immediate and incurs additional costs. This solution is suitable for archival data rather than frequent access, making it less ideal for the application’s requirements.
I plan that the answer is:
Create an Amazon S3 bucket that uses an Intelligent-Tiering lifecycle policy. Copy all files to the S3 bucket. Update the application to use Amazon S3 API to store and retrieve files.