In addition to Amazon S3 cloud, there is some other storage services provided by Amazon, such as Glacier storage for long-term keeping of those different data which are not entered into a day-by-day workflow.
For these “preserved” data, Amazon Glacier backup is a good method of keeping intact.
Amazon Glacier Backup Step by Step
1. Prepare Your AWS Environment
- Sign in to the AWS Management Console.
- Make sure you have an S3 bucket. If not, create one via S3 → Create bucket.
- Choose a bucket name (globally unique) and region close to you.
2. Upload Files Directly to Glacier Storage Classes
Option A: Using the AWS S3 Console
- Open your bucket → click Upload.
- Add files or folders.
- Under Properties → Storage Class, select either:
- S3 Glacier Instant Retrieval – nearline storage, fast access, slightly higher cost.
- S3 Glacier Flexible Retrieval – cheaper storage, retrieval takes minutes to hours.
- Complete upload. Files are now stored in the chosen Glacier tier.
Option B: Using AWS CLI
Upload with a specific storage class:
aws s3 cp myfile.zip s3://my-backup-bucket/ --storage-class GLACIER_IR(for Instant Retrieval)
aws s3 cp myfile.zip s3://my-backup-bucket/ --storage-class GLACIER(for Flexible Retrieval)
3. Automate with Lifecycle Rules
Instead of choosing Glacier every time, you can set rules:
- Go to S3 → Bucket → Management → Lifecycle rules.
- Create a new rule:
- Transition objects to Glacier Instant Retrieval after, e.g., 30 days.
- Transition to Glacier Flexible Retrieval after 90 days.
- This ensures new files automatically move to cheaper storage without manual uploads.
4. Retrieve Your Glacier Backed Up Data
- Instant Retrieval: Access objects almost immediately (milliseconds).
- Flexible Retrieval: Requires restore request.
Restore Glacier backup with Console:
- Select the object in your S3 bucket.
- Click Initiate Restore.
- Choose Retrieval tier:
- Standard (3–5 hours)
- Expedited (1–5 minutes, higher cost)
- Bulk (5–12 hours, cheapest)
- Once restored, download as usual.
Restore Glacier backup with CLI:
aws s3api restore-object --bucket my-backup-bucket --key myfile.zip --restore-request '{"Days":1,"GlacierJobParameters":{"Tier":"Standard"}}'
This makes the file temporarily available for download.
AWS Glacier Backup Cost and Other Considerations
- AWS Glacier Storage Costs: Glacier Deep Archive is cheapest per-GB; Flexible Retrieval is next; Instant Retrieval is slightly higher (though still much lower than Standard). You can save up to ~95% versus standard S3 by using Deep Archive for very long-lived data. Remember that each restore (and each lifecycle transition) is an S3 API request, which has a small fee.
- Size Thresholds: Keep in mind the 128 KB minimum charge for Instant Retrieval. If you archive many tiny files, most of their bytes go to metadata overhead. Consider bundling small files or filtering them out.
- Metadata Overhead: Each object in Flexible or Deep incurs ~40 KB of extra storage for S3’s indexing (charged as 8 KB at Standard rate + 32 KB at Glacier rate). For large backups of many files, this can add up; archiving a few huge files is often more cost-efficient.
- Glacier Backup Versioning: If enabled, every version of an object transitions and incurs cost independently. Old versions remain archived unless you expire them with lifecycle rules. Restoring without a version ID always restores the latest version.
- Data Access Patterns: Glacier classes are ideal for immutable backups, archives, and DR copies. They are not meant for data that you may need frequently or unpredictably. For mixed or automated archiving, consider S3 Intelligent-Tiering which can auto-move data between tiers including Glacier Deep Archive without retrieval charge.
- Lifecycle Propagation: After enabling a lifecycle rule, S3 may take a few minutes to propagate the config. It also treats the transition schedule loosely (asynchronously), so objects might transition a few days after the rule’s threshold. Don’t be alarmed if a “30-day” rule actually runs on day 31 or 32 for some objects.
AWS Glacier Backup Practical Tips from User Experience
- Test with Small Objects: Try a small test file and archive it to each class to see how the process works end-to-end. Note that if it’s under 128 KB, S3 will still charge for 128 KB in Glacier Instant Retrieval.
- Use Proper CLI Versions: Ensure your AWS CLI is up-to-date (v2.4.5 or later) to support GLACIER_IR. In older CLI versions you may only be able to use s3api put-object (not s3 cp) for the Instant class.
- Monitor Lifecycle Results: After enabling a rule, use the S3 console or aws s3api list-objects to check the storage class of objects. Remember that objects do not “immediately” vanish from Standard; they only switch classes after the specified age.
- Beware of Separate Glacier Service: Since these objects remain in S3, you cannot manage them with the old Glacier vault APIs or console. All operations (listing, restoring, deleting) must be done via S3’s APIs or console.
- Expiration vs Transition: If you set both a transition and an expiration rule, note that expiration (deletion) takes precedence. An object scheduled to transition on day 100 but expire on day 90 will be deleted before being archived.
- The Glacier Batch Operations: For bulk restores (e.g. hundreds of files), consider S3 Batch Operations to script restore requests. This avoids manually clicking each object.
By following these steps and precautions, individual users and small teams can safely archive backups to Glacier classes using only AWS-native tools. You gain the same durability benefits of S3 at a fraction of the storage cost, while being able to retrieve your Glacier backup data when needed – just be mindful of the minimum sizes, costs, and delays involved.
For those who want to learn about how to backup Amazon S3 services we recommend reading our detailed guide.
How To Backup S3 With External Backup Tool
Looking for a simple yet powerful way to protect your data without any IT expertise? Discover Handy Backup — the ultimate automatic S3 backup solution. Enjoy effortless multi-cloud backup, with seamless support for Dropbox, OneDrive, and many other cloud services — all in one easy-to-use tool.
Selection of Amazon Glacier Backup Tool Software
These are some featured Amazon Glacier backup client software. If you will plan to store on Amazon Glacier Windows Server backup or a database snapshot, then CloudBerry will help. For home Amazon Glacier photo backup or video archiving, this is freeware FastGlacier to use.
Note that these solutions often have only a limited capability to store different types of crucial data, making Amazon Glacier incremental backup, encrypting data snapshots and so on. For a full-featured Amazon Glacier backup utility, please refer to future Handy Backup plug-in!