A company is running a development application on an Amazon EC2 instance. The application uploads 500,000 files that are 1 GB in size into a target Amazon S3 bucket that has default encryption…

Question

Options

• AEnable S3 Transfer Acceleration on the S3 bucket.
• BSplit the S3 write operations to use multiple bucket prefixes to write items in parallel.
• CConfigure AWS PrivateLink for Amazon S3. Turn off encryption on the S3 bucket.
• DConfigure AWS Global Accelerator in the Region. Turn off encryption on the S3 bucket.

• A
  12% (3)
• B
  80% (20)
• C
  4% (1)
• D
  4% (1)

Explanation

Option B is correct because Amazon S3 scales request throughput per prefix - up to 3,500 PUT requests per second per prefix. With 500,000 large files being written sequentially to the same prefix, you hit this rate limit, causing the application to wait. Distributing writes across multiple prefixes (e.g., using hash-based prefixes like a/, b/, c/) multiplies available throughput linearly.

Why the distractors are wrong:

• A (Transfer Acceleration) routes traffic through CloudFront edge nodes to reduce latency over long distances - it provides zero benefit when EC2 and S3 are in the same Region.
• C (PrivateLink + disable encryption) - PrivateLink keeps traffic on the AWS private network, but since networking is already confirmed non-constrained, this changes nothing; disabling encryption offers negligible throughput gain and is a security regression.
• D (Global Accelerator + disable encryption) - Global Accelerator optimizes routing for internet-facing TCP/UDP traffic, not internal S3 API calls; it has no effect on same-region S3 writes.

Memory tip: Think of S3 prefixes as checkout lanes at a grocery store - one lane for 500,000 customers causes a massive queue, but opening more lanes (prefixes) lets customers (requests) flow through in parallel. The key phrase on the exam is "write in parallel" → multiple prefixes.

No community discussion yet for this question.