AWS Certified Solutions Architect – ProfessionalAccelerate Workload Migration and ModernizationHard

A global media company is migrating a large video archive (hundreds of petabytes) from on-premises tape libraries to AWS. The archive is rarely accessed, but when it is, retrieval must be possible within a few hours. Cost optimization for storage is the primary concern, followed by reliable retrieval. Which AWS storage service and strategy should be employed?

  1. AAmazon S3 Glacier Instant Retrieval for rapid access to archived data.
  2. BAmazon S3 Glacier Deep Archive with S3 Lifecycle policies.
  3. CAmazon S3 Standard for immediate access.
  4. DAmazon FSx for Lustre for high-performance file access.
Show answer & explanation

Correct answer: B. Amazon S3 Glacier Deep Archive with S3 Lifecycle policies.

Amazon S3 Glacier Deep Archive is the most cost-effective storage class for long-term archiving of infrequently accessed data (hundreds of petabytes). With retrieval times ranging from 12 hours, it meets the 'within a few hours' requirement for rarely accessed data when considering the extreme cost savings. S3 Lifecycle policies can automate transitions if data retention requirements change.

Why the other options are wrong

  • A. S3 Glacier Instant Retrieval offers millisecond retrieval, but at a higher cost than Deep Archive, which is not the primary concern. For 'rarely accessed' and 'cost optimization primary', Deep Archive is more suitable.
  • C. S3 Standard is too expensive for rarely accessed, long-term archives of hundreds of petabytes.
  • D. Amazon FSx for Lustre is a high-performance file system for compute-intensive workloads, not suitable or cost-effective for long-term, rarely accessed archiving.

Amazon S3 Glacier Storage Classes

A family of Amazon S3 storage classes optimized for archiving data that is infrequently accessed, offering various retrieval times and cost points.

  • Includes S3 Glacier Instant Retrieval, S3 Glacier Flexible Retrieval, and S3 Glacier Deep Archive.
  • Designed for cost-effective long-term data retention.
  • Different classes offer trade-offs between retrieval speed and cost.

Memory trick: Cost vs. Time: Standard, IA, Glacier, Deep Archive.

More Accelerate Workload Migration and Modernization questions