Course menuSAP-C03

Exams / SAP-C03 / Study guide / Storage tiering and volume costs (S3 classes, lifecycle, gp3)

Storage tiering and volume costs (S3 classes, lifecycle, gp3): SAP-C03 study notes and practice questions

What to know about Storage tiering and volume costs (S3 classes, lifecycle, gp3) for the SAP-C03 exam, from the Cost-Optimized Architecture Design domain (15% of the exam guide), then 1 free question to test it. Part of the SAP-C03 study guide.

Last verified against current AWS documentation ·How we write our questions

What to know for the exam

SAP-C03 gives you a storage bill, an access pattern (“read often for 30 days, rarely after”, “unpredictable”, “must keep 7 years”) and a retrieval need, and asks for the cheapest design that still meets it.

S3 storage classes

Class Fits Watch for
S3 Standard Frequently accessed data Highest storage price
S3 Intelligent-Tiering Unknown or changing access patterns Small monitoring fee per object; objects under 128 KB are not monitored or tiered; no retrieval fees
S3 Standard-IA Infrequent access, needs millisecond reads 30-day minimum, 128 KB minimum billable size, per-GB retrieval fee
S3 One Zone-IA Re-creatable infrequent data Stored in one AZ only
S3 Glacier Instant Retrieval Accessed about once a quarter, needs milliseconds 90-day minimum
S3 Glacier Flexible Retrieval Archives, minutes to hours to restore 90-day minimum
S3 Glacier Deep Archive Long-term retention, rarely if ever read 180-day minimum; standard retrieval within 12 hours
  • Lifecycle rules transition objects by age (and by prefix, tag or size filter) and expire objects and noncurrent versions. Add a rule to abort incomplete multipart uploads and expire old delete markers.
  • Use lifecycle rules when the pattern is known and age-based; use Intelligent-Tiering when it isn’t. Intelligent-Tiering can also move objects to optional Archive Access and Deep Archive Access tiers after long inactivity.
  • S3 Storage Lens and storage class analysis show which prefixes are cold before you write rules.
  • Many tiny objects cost more to transition than they save; aggregate them or keep them in Standard.

Amazon EFS and FSx

  • EFS lifecycle management moves files not accessed for a set period to Infrequent Access and then Archive, and can move them back to Standard on first access.
  • EFS One Zone storage costs less for data that doesn’t need multi-AZ durability.
  • FSx file systems can use data deduplication (FSx for Windows File Server and FSx for ONTAP), compression (FSx for ONTAP) and capacity pool tiering (FSx for ONTAP, FSx for OpenZFS intelligent tiering) to cut cost.

Amazon EBS

Type Facts
gp3 Baseline 3,000 IOPS and 125 MiB/s at any size; provision IOPS (up to 80,000) and throughput (up to 2,000 MiB/s) separately from size; up to 64 TiB; lower price per GB than gp2
gp2 IOPS tied to size (3 IOPS per GiB, burst to 3,000 for small volumes); you often overbuy size to get IOPS
io2 Block Express Sub-millisecond latency, up to 256,000 IOPS, highest durability; for critical databases
st1 / sc1 Throughput or cold HDD for large sequential data
  • Migrate gp2 to gp3 in place with Elastic Volumes, no downtime. Many io1 volumes that need only a few thousand IOPS also fit gp3 for less.
  • Delete unattached volumes and old snapshots; Compute Optimizer recommends EBS volume changes.

How to choose

Scenario wording Pick
“Access pattern unknown, no retrieval fees wanted” S3 Intelligent-Tiering
“Logs read for 30 days, kept 7 years, rarely read” Lifecycle to Glacier Deep Archive, expire at 7 years
“Medical images read a few times a year, need instant access” S3 Glacier Instant Retrieval
“Large gp2 volumes created only for IOPS” gp3 with provisioned IOPS
“Shared file data mostly cold after 30 days” EFS lifecycle to IA and Archive

Common traps

  • Moving objects to Standard-IA or Glacier and deleting them before the minimum duration, which still bills the full minimum.
  • Lifecycle transitions for millions of tiny objects.
  • Glacier Flexible Retrieval when the requirement says millisecond access.
  • io2 for a general workload that gp3 serves.
  • Forgetting noncurrent versions in versioned buckets.

Try the 1 practice question

A media company keeps a 300 TB archive of video files in the Amazon S3 Standard storage class. Some files are watched every week, many are not opened for months, and the access pattern cannot be predicted per file. The archive grows by about 10 TB each month.

The team wants the storage cost to fall without a person reviewing each file, and it does not want to pay a surprise retrieval charge when a rarely used file is opened.

Which storage approach meets these requirements?

Choose one answer.

  1. A. Move every object to the S3 Glacier Deep Archive storage class with a lifecycle rule, so that the whole archive is billed at the lowest storage rate that Amazon S3 offers.
  2. B. Keep the objects in S3 Standard and rely on the team to move the files it knows are old to a cheaper class each quarter.
  3. C. Move every object to the S3 One Zone-IA storage class, which costs less than Standard for all files.
  4. D. Store the archive in S3 Intelligent-Tiering, which moves each object between access tiers based on its use, with no retrieval charges in the frequent and infrequent tiers.
Show the answer and explanation

Answer: D

  • A. Wrong: Deep Archive has long retrieval times and retrieval charges, and the weekly-viewed files would be slow and costly to open. The unpredictable pattern makes this a poor fit.
  • B. Wrong: A manual quarterly review needs someone to review each file, which the team wants to avoid. It also leaves the storage at the Standard rate for the files nobody reviews.
  • C. Wrong: One Zone-IA stores data in a single Availability Zone and charges retrieval fees, so the archive loses resilience and the opened files cost more. It does not suit the unpredictable pattern.
  • D. Right: Correct. Intelligent-Tiering moves each object automatically based on observed access, with no per-file review. The frequent and infrequent access tiers have no retrieval charges, so opening an older file does not add a surprise fee.

Explanation

Intelligent-Tiering moves each object based on its own access pattern with no per-file review and no retrieval fees in its frequent and infrequent tiers. Deep Archive and One Zone-IA add retrieval costs or lose resilience, and Standard with manual moves leaves cost on the table.

Study

Want more on Storage tiering and volume costs (S3 classes, lifecycle, gp3)? The 5 full-length practice sets have 5 more Storage tiering and volume costs (S3 classes, lifecycle, gp3) questions, 375 in all, none of them in the free exam. One payment of $14.99, no subscription. See the practice sets

Study Storage tiering and volume costs (S3 classes, lifecycle, gp3)

More Cost-Optimized Architecture Design topics