MLA-C01 Practice Question 71: On Premises Data Transfer
AWS Certified Machine Learning Engineer – Associate (MLA-C01) · Test 8 · question 1 of 10
A research lab keeps a 40 TB microscopy image dataset on an on-premises NFS file server, which stays the lab's primary copy. The lab wants to train models on SageMaker AI from Amazon S3. It needs an initial full copy followed by a weekly copy of only new and changed files, sent over its existing AWS Direct Connect link, with data integrity verified after each transfer and without writing or maintaining transfer scripts.
Which service should the lab use?
Show answer and explanation
Answer:
AWS DataSync is a managed transfer service that moves data between on-premises storage, including NFS file servers, and AWS storage services such as Amazon S3. A DataSync agent reads the NFS share and a task copies the data to S3. Tasks can run on a schedule, transfer only what has changed since the last run, and verify data integrity. That covers the initial copy, the weekly copies, and the verification with no scripts to maintain (option B).
- A is wrong because it is exactly the scripting the lab wants to avoid. The team would own the cron job, retries, logging, and any integrity checking beyond what the CLI does on its own. - C is wrong because Snowball Edge is for moving data where the network is too slow or absent. The lab has Direct Connect, and shipping devices every week adds days of delay for no benefit. - D is wrong because a File Gateway gives on-premises applications a file interface backed by S3. Someone would still have to copy the images onto that share and track what changed, which is the transfer job DataSync already does.
Reference: What is AWS DataSync?