AWS Certified Solutions Architect – ProfessionalDesign Solutions for Organizational ComplexityHard

A global media company is migrating its video transcoding pipeline to AWS. The existing on-premises solution uses a custom batch processing system that generates hundreds of terabytes of intermediate files during the transcoding process. These files require high-throughput, low-latency access from compute instances for sequential read/write operations and must be shared across thousands of parallel transcoding jobs. The company wants to optimize for performance and cost while leveraging a fully managed service. Which AWS storage solution should be recommended?

  1. AAmazon EBS Provisioned IOPS volumes attached to individual EC2 instances.
  2. BAmazon S3 Standard with lifecycle policies to move data to S3 Standard-IA.
  3. CAmazon FSx for Lustre integrated with Amazon S3.
  4. DAmazon EFS configured for general purpose performance mode.
Show answer & explanation

Correct answer: C. Amazon FSx for Lustre integrated with Amazon S3.

Amazon FSx for Lustre is specifically designed for high-performance computing (HPC) workloads like video transcoding, offering very high throughput and low-latency access for sequential operations across thousands of parallel jobs. Its integration with Amazon S3 allows for cost-effective long-term storage of source/final assets while providing a high-performance scratch space for intermediate files, making it a fully managed and performant solution.

Why the other options are wrong

  • A. Amazon EBS volumes are block storage attached to a single EC2 instance. This would require complex distributed file system software to share data across thousands of instances, defeating the purpose of a fully managed service and introducing significant management overhead and latency for shared access.
  • B. Amazon S3 is object storage and while highly scalable, it does not provide the file system semantics, low-latency, or high-throughput shared access required for intermediate files in a high-performance transcoding pipeline. It's better for source/final assets.
  • D. Amazon EFS is a shared file system, but its performance characteristics (especially for throughput-intensive, sequential I/O across thousands of instances) are generally lower than FSx for Lustre and it might not meet the extreme performance demands of large-scale video transcoding.

Amazon FSx for Lustre

A fully managed file system optimized for compute-intensive workloads, such as high-performance computing (HPC), machine learning, media processing, and financial simulations.

  • Provides very high throughput and low-latency access.
  • Designed for shared access across thousands of compute instances.
  • Integrates with Amazon S3 for data persistence and cost-effectiveness.

Memory trick: Lustre is for lightning-fast, shared storage needs.

More Design Solutions for Organizational Complexity questions