AWS Certified Solutions Architect – Associate (SAA-C03) — Question 651

A company regularly uploads GB-sized files to Amazon S3. After the company uploads the files, the company uses a fleet of Amazon EC2 Spot Instances to transcode the file format. The company needs to scale throughput when the company uploads data from the on-premises data center to Amazon S3 and when the company downloads data from Amazon S3 to the EC2 instances.

Which solutions will meet these requirements? (Choose two.)

Answer options

• A. Use the S3 bucket access point instead of accessing the S3 bucket directly.
• B. Upload the files into multiple S3 buckets.
• C. Use S3 multipart uploads.
• D. Fetch multiple byte-ranges of an object in parallel.
• E. Add a random prefix to each object when uploading the files.

Correct answer: C, D

Explanation

S3 multipart uploads improve upload throughput by allowing large files to be broken down and uploaded in parallel streams. For downloading, fetching multiple byte-ranges of an object concurrently allows the Amazon EC2 instances to parallelize the download process, which significantly increases read performance. Other options like using S3 access points, multiple buckets, or random prefixes do not inherently scale the parallel throughput of individual large file transfers.