Opportunity Name

S3 Glacier to Deep Archive

AWS Resource Type (AWS service name)

Amazon Simple Storage Service (Amazon S3)

Opportunity Description

This Finder identifies Amazon S3 buckets that currently store data in S3 Glacier Flexible Retrieval but do not have a lifecycle rule to transition eligible objects to S3 Glacier Deep Archive.

The goal is to help reduce long-term archival storage costs for data that is rarely or never accessed, such as dormant backups, compliance archives, and old snapshots. Based on the source material, Glacier Flexible Retrieval is priced at $0.004 per GB/month and Glacier Deep Archive at $0.00099 per GB/month, representing an estimated 75.25% reduction for suitable archival data.

CloudFix evaluates bucket-level Glacier storage costs, checks whether a Deep Archive lifecycle transition already exists, and estimates the annual savings from adding a transition rule.

Criteria for identifying the opportunity

CloudFix identifies this opportunity when all of the following are true:

• The S3 bucket has Glacier storage costs visible in CUR data.

• The bucket does not already have a lifecycle rule that transitions objects to DEEP_ARCHIVE.

• The bucket is still present and accessible during validation.

• There is no conflicting expiration rule that would delete objects before the configured Deep Archive transition threshold.

• The projected annual savings exceed the configured minimum savings threshold.

The Finder is designed around these checks:

1. CUR-based candidate discovery
   CloudFix queries CUR data to find S3 buckets with Glacier storage usage and cost.

2. Bucket existence validation
   CloudFix verifies the bucket still exists using S3 validation.

3. Lifecycle validation
   CloudFix checks the bucket lifecycle configuration and skips buckets that already have a Deep Archive transition rule.

4. Savings threshold validation
   CloudFix only creates a recommendation if the estimated annual savings are high enough to be actionable.

Potential Savings (if known)

Based on the source material, Deep Archive is estimated to be 75.25% cheaper than Glacier Flexible Retrieval for storage.

CloudFix calculates savings using actual Glacier storage cost from CUR data:

• Projected Annual Glacier Cost = current Glacier cost annualized from CUR data

• Projected Annual Deep Archive Cost = projected annual Glacier cost × 0.2475

• Potential Annual Savings = projected annual Glacier cost × 0.7525

Using the formula from the specification:

Potential Annual Savings = Current Glacier Cost × 0.7525

This Finder generates one recommendation per bucket, so customers can prioritize the highest-value opportunities first.

What happens when the Fixer is Executed?

There is no automatic Fixer in the initial release.

Instead, CloudFix generates a recommendation that includes:

• the bucket name and region,

• current Glacier cost,

• projected Deep Archive cost,

• estimated annual savings,

• a lifecycle rule template,

• and CLI guidance for remediation.

The customer must implement the lifecycle rule manually in AWS.

Once the lifecycle rule is applied, AWS lifecycle processing can transition eligible existing Glacier objects to Deep Archive automatically, subject to the configured transition threshold and AWS lifecycle behavior.

Is it possible to roll back once CloudFix implements the Fixer?

CloudFix does not apply this change automatically, so there is no CloudFix rollback action.

If a customer manually changes lifecycle rules in AWS, rollback would also be manual. The customer could modify or remove the lifecycle rule, but any objects already transitioned to Deep Archive would remain in that storage class unless restored or otherwise handled through AWS processes.

Can CloudFix implement the fix automatically once I accept the recommendation?

No. This is currently a manual remediation opportunity.

CloudFix identifies the savings opportunity and provides implementation guidance, but the lifecycle rule must be added by the customer in AWS.

Does the fix require downtime?

No. This is a non-destructive storage class transition and does not require application downtime.

The optimization changes the storage class of eligible objects; it does not delete the data or alter bucket availability. Customers should still review retrieval-time trade-offs before implementing the recommendation.

Notes for customers

This recommendation is best suited to data that is rarely accessed. The source material highlights these operational considerations:

• Deep Archive retrieval is slower than Glacier Flexible Retrieval.

• Deep Archive has a 180-day minimum retention consideration.

• This is intended for long-term archival data, not data with frequent restore requirements.

• The recommendation is non-destructive, but customers should confirm retrieval expectations, retention obligations, and compliance requirements before implementation.