Community hub
0 subscribers7 pages, 0 posts
Recent from talks
All channels
Be the first to start a discussion here.
Be the first to start a discussion here.
Be the first to start a discussion here.
Be the first to start a discussion here.
Contribute something
Welcome to the community hub built to collect knowledge and have discussions related to Amazon S3 Glacier.
Nothing was collected or created yet.
Amazon S3 Glacier
View on Wikipediafrom Wikipedia
Amazon S3 Glacier is an online file storage web service that provides storage for data archiving and backup.[2]
Key Information
Glacier is part of the Amazon Web Services suite of cloud computing services, and is designed for long-term storage of data that is infrequently accessed and for which retrieval latency times of 3 to 5 hours are acceptable. Storage costs are a consistent $0.004 per gigabyte per month, which is substantially cheaper than the Simple Storage Service (S3) Standard tier.[3]
Amazon hopes this service will move businesses from on-premises tape backup drives to cloud-based backup storage.[4]
Storage
[edit]The underlying technology used by Glacier is unknown and subject to speculation.
Amazon officially states in their S3 FAQS:[5]
Q: What is the backend infrastructure supporting the S3 Glacier Flexible Retrieval and S3 Glacier Deep Archive storage class?
We prefer to focus on the customer outcomes of performance, durability, availability, and security. However, this question is often asked by our customers. We use a number of different technologies which allow us to offer the prices we do to our customers. Our services are built using common data storage technologies specifically assembled into purpose-built, cost-optimized systems using AWS-developed software. The S3 Glacier storage classes benefit from our ability to optimize the sequence of inputs and outputs to maximize efficiency accessing the underlying storage.
ZDNet says, that according to private e-mail, Glacier runs on "inexpensive commodity hardware components".[4] In 2012, ZDNet quoted a former Amazon employee as saying that Glacier is based on custom low-RPM hard drives attached to custom logic boards where only a percentage of a rack's drives can be spun at full speed at any one time.[6][7] Similar technology is also used by Facebook.[8]
There is some belief among users that the underlying hardware used for Glacier storage is tape-based, because Amazon has positioned Glacier as a direct competitor to tape backup services (both on-premises and cloud-based).[9] This confusion is exacerbated by the fact that Glacier has archive retrieval delays (3–5 hours before archives are available) similar to that of tape-based systems[dubious – discuss] and a pricing model that discourages frequent data retrieval.[10]
The Register claimed that Glacier runs on Spectra T-Finity tape libraries with LTO-6 tapes.[11][12] Others have conjectured Amazon using off-line shingled magnetic recording hard drives, multi-layer Blu-ray optical discs, or an alternative proprietary storage technology.[13]
Data storage consultant Robin Harris speculated that the storage is based on cheap optical disks such as Blu-ray, based on hints from public sources.[14]
Cost
[edit]Glacier has two costs, one for storage and one for retrieval. Uploading data to Glacier is free. Storage pricing is simple: it currently costs 0.4 cents per gigabyte per month, which is 82% cheaper than S3 Standard. When Glacier launched in 2012, the storage charge was set to 1 cent per gigabyte per month. This was reduced to 0.7 cents in September 2015 and to the current 0.4 cents in December 2016.[15]
Glacier used to charge for retrievals based on peak monthly retrieval rate, meaning that (ignoring the free tier) if you downloaded four gigabytes in four hours, it would cost the same as if you downloaded 720 gigabytes in 720 hours, in a 30-day month. This made it cheaper to spread out data retrievals over a long period of time, but failing to do so could result in a surprisingly large bill. In one case, a user stored 15 GB of data in Glacier, retrieved 693 MB for testing, and ended up being charged for 126 GB due to retrieval rate calculation.[16] This pricing policy was widely regarded as a "time bomb" set to go off on retrieval.[17]
In 2016, AWS revised their retrieval pricing model.[18] The new model bases the retrieval fee on the number of gigabytes retrieved. This can amount to a 99% price cut for users who perform only one Glacier retrieval in a month. At the same time, AWS introduced new methods of retrieval that take different amounts of time. An expedited retrieval costs one cent per request and three cents per gigabyte, and can retrieve data in one to five minutes. A standard retrieval costs five cents per thousand requests and one cent per gigabyte, and takes three to five hours. A bulk retrieval costs 2.5 cents per thousand requests and 0.25 cents per gigabyte, and takes seven to twelve hours. AWS also introduced provisioned capacity for expedited retrievals, each unit of which costs $100 per month and guarantees at least three expedited retrievals every five minutes, and up to 150 MB/s of retrieval bandwidth. Without provisioned capacity, expedited retrievals are done on a capacity available basis.[19]
Data deleted from Glacier less than 90 days after being stored incurs a charge equal to the cost of storage for the remainder of the 90 days. (In effect, the user pays for 90 days minimum.) This move was designed to discourage the service's use in cases where Amazon's other storage offerings (e.g. S3) are more appropriate for real-time access. After 90 days, deletion from Glacier is free.
Retrieving data from Glacier is a two-step process. The first step is to retrieve the data into a staging area, where it stays for 24 hours.[20] The second step is to download the data from the staging area, which may incur bandwidth charges.[21]
Glacier is also available as a "storage class" in S3.[22] Objects can only be put into Glacier by lifecycle rules, which can be configured to put the objects in Glacier once they have reached a certain age. Pricing is the same, but there is no staging area; instead, retrieved objects are simultaneously stored in Glacier and in Reduced Redundancy class for a number of days that the user specifies.
References
[edit]- ^ Jeff Barr (August 21, 2012). "Amazon Glacier: Archival Storage for One Penny Per GB Per Month". AWS Blog. Retrieved November 29, 2016.
- ^ Mlot, Stephanie (August 21, 2012). "Amazon Launches Glacier Cloud Storage Service". PCMag.com. Ziff Davis, Inc. Retrieved August 21, 2012.
- ^ "Pricing". Aws.amazon.com. Retrieved June 18, 2015.
- ^ a b Clark, Jack (August 21, 2012). "Amazon launches Glacier cloud storage, hopes enterprise will go cold on tape use". ZDNet. CBS Interactive. Retrieved August 21, 2012.
- ^ "Amazon Simple Storage Service (S3) — Cloud Storage — AWS". Amazon Web Services, Inc. Retrieved 2022-09-09.
- ^ Clark, Jack (August 24, 2012). "Could the tech beneath Amazon's Glacier revolutionise data storage?". ZDNet. Retrieved June 18, 2015.
- ^ "Former S3 employee here. I was on my way out of the company just after the stora... | Hacker News". News.ycombinator.com. Retrieved June 18, 2015.
- ^ Gallagher, Sean (November 9, 2015). "How Facebook puts petabytes of old cat pix on ice in the name of sustainability". Ars Technica.
- ^ "Amazon Glacier: 99.999999999% durability long-term storage, for a penny a gig". ExtremeTech. August 21, 2012. Retrieved June 18, 2015.
- ^ Paul Cooper (November 9, 2013). "One of tech's most elusive mysteries: The secret of Amazon Glacier". IT ProPortal.
- ^ "Insider 'fesses up: Amazon's Glacier cloud is made of ... TAPE". Theregister.co.uk. Retrieved June 18, 2015.
- ^ "Spectra: Tape is dead? We installed 550PB of the stuff in 6 months". Theregister.co.uk. Retrieved June 18, 2015.
- ^ "Amazon's Glacier secret: BDXL". Storagemojo.com. Retrieved June 18, 2015.
- ^ Harris, Robin. "Amazon's Glacier secret: BDXL | StorageMojo".
- ^ "The cloud price war continues: Amazon cuts its cloud storage prices, again". ZDNet. Retrieved September 4, 2019.
- ^ "FastGlacier surprising Retrieval Fee". AWS Developer Forums. Aws.amazon.com. September 21, 2012. Retrieved January 30, 2013.
- ^ Finley, Klint (August 21, 2012). "Is There a Landmine Hidden in Amazon's Glacier?". Wired – via www.wired.com.
- ^ "AWS Storage Update – S3 & Glacier Price Reductions + Additional Retrieval Options for Glacier". aws.amazon.com. November 21, 2016. Retrieved February 1, 2018.
- ^ "Glacier FAQ: Data Retrievals". aws.amazon.com. Retrieved February 1, 2018.
- ^ "Retrieving Amazon Glacier Archives". aws.amazon.com. Retrieved February 1, 2018.
- ^ "Amazon Glacier Pricing". aws.amazon.com. Retrieved February 1, 2018.
- ^ "Amazon S3 Storage Classes". aws.amazon.com. Retrieved February 1, 2018.
External links
[edit]
Amazon S3 Glacier
View on Grokipediafrom Grokipedia
Overview
Introduction
Amazon S3 Glacier is a low-cost object storage service within Amazon Web Services (AWS) Simple Storage Service (S3), designed specifically for data archiving and long-term backup of infrequently accessed information.[1] It enables users to securely store large volumes of data, such as backups, media archives, and compliance records, while providing virtually unlimited scalability.[2] The service emphasizes exceptional durability, engineered to deliver 99.999999999% (11 nines) data durability over a given year through redundant storage across multiple devices and facilities.[1] It also offers high availability tailored to archival needs, ensuring reliable access when data is retrieved.[2] Unlike standard S3 storage classes optimized for frequent access, Amazon S3 Glacier prioritizes cost efficiency for data that is rarely retrieved, with restoration times typically spanning from minutes to hours or days rather than milliseconds.[1] This makes it ideal for scenarios where immediate access is not required, such as long-term retention for regulatory compliance or disaster recovery.[2] Launched in 2012 as part of the AWS S3 ecosystem, it has evolved to include multiple storage classes by 2025, allowing flexible options for varying archival requirements.[3] At its core, Amazon S3 Glacier stores data as objects within S3 buckets, supporting automatic tiering through lifecycle policies to transition objects between storage classes based on access patterns and retention needs.[2] This integration with S3 provides seamless management, including features like versioning, encryption, and access controls, while maintaining the service's focus on economical, durable archival storage.[1]History
Amazon S3 Glacier was initially launched on August 20, 2012, as a standalone archival storage service by Amazon Web Services (AWS), designed for secure, durable, and low-cost data archiving at $0.01 per GB per month, utilizing a vault-based architecture with retrieval times of 3 to 5 hours. This service addressed the need for affordable long-term storage separate from the core Amazon S3 object storage, enabling customers to archive data without the higher costs of frequent access tiers. In November 2012, AWS introduced integration with Amazon S3 Lifecycle policies, allowing automatic transitions of S3 objects to Glacier vaults for cost optimization. By November 2016, AWS enhanced Glacier with significant price reductions—to $0.004 per GB per month in key regions—and expanded retrieval options, including expedited (1-5 minutes) and bulk (5-12 hours) tiers to improve flexibility for varying access needs.[5] The service saw further evolution in November 2018 when AWS officially integrated it into Amazon S3 as a native storage class, renaming it Amazon S3 Glacier and enabling direct uploads via S3 APIs, which simplified management and eliminated the need for separate vault operations.[9] This transition marked a shift toward seamless incorporation within the S3 ecosystem, supporting broader adoption for archival workloads. In March 2019, AWS introduced the S3 Glacier Deep Archive storage class, offering even lower costs at $0.00099 per GB per month for data accessed less than once or twice a year, with retrieval times of 12 to 48 hours, targeting ultra-long-term retention scenarios.[10] Advancements continued in November 2021 with the launch of S3 Glacier Instant Retrieval, providing millisecond access for rarely retrieved data at $0.004 per GB per month, alongside renaming the original S3 Glacier class to S3 Glacier Flexible Retrieval to better reflect its variable retrieval options and include free bulk retrievals.[11] As of 2025, AWS announced that the original standalone Amazon Glacier service would cease accepting new customers starting December 15, 2025, fully directing users to the integrated S3 Glacier storage classes, with no disruptions for existing customers and continued support for hybrid cloud integrations via enhanced APIs like AWS DataSync.[12] This update underscores the complete migration to S3-native archival storage, with minor ongoing optimizations in pricing and API capabilities but no new storage classes introduced.Storage Classes
Glacier Instant Retrieval
Amazon S3 Glacier Instant Retrieval is designed for long-lived data that is rarely accessed, typically once per quarter, while providing millisecond retrieval times to support performance-sensitive archival needs.[13] This storage class offers real-time access with the same latency and throughput as S3 Standard-Infrequent Access, enabling near-instant retrieval for objects without the need for restoration processes.[2] It was introduced on November 30, 2021, as a new option within the S3 Glacier family to bridge the gap between frequent-access storage classes and traditional archival tiers, allowing users to optimize costs for data that requires occasional but rapid recovery.[11] Key specifications include a minimum storage duration of 90 days, after which early deletion incurs pro-rated charges for the full period.[7] For billing purposes, objects smaller than 128 KB are charged at the 128 KB rate to account for the fixed overhead in archival storage.[2] The class provides high durability of 99.999999999% (11 nines) over a given year, ensuring robust protection against data loss, and an availability of 99.9% over the same period. This storage class is ideal for active archives where quick recovery is essential, such as media assets, user-generated content in online file-sharing platforms, medical imaging, genomics data, and health records.[13] It balances cost efficiency with accessibility, making it suitable for scenarios where data is not frequently retrieved but must be available without significant delays, unlike slower archival options.[1] Trade-offs include higher storage costs compared to deeper archive classes, offset by the elimination of retrieval fees and expedited access for time-critical applications.[2]Glacier Flexible Retrieval
Amazon S3 Glacier Flexible Retrieval, formerly known as S3 Glacier, is an archival storage class designed for data that requires occasional access with retrieval times measured in minutes to hours.[14] It provides three retrieval options to balance speed and cost: expedited retrievals complete in 1–5 minutes for urgent needs, standard retrievals typically finish within 3–5 hours, and bulk retrievals take 5–12 hours at no additional retrieval cost.[15] These options allow users to provision capacity for faster restores when necessary, making it suitable for scenarios where immediate access is not required but flexibility is valued.[2] Introduced as part of the original S3 Glacier service and renamed in November 2021 to emphasize its retrieval versatility, this class features a minimum storage duration of 90 days and imposes no minimum object size, enabling efficient storage of diverse data sets from small files to large archives.[14] Upon renaming, storage costs were reduced by up to 10% compared to the prior S3 Glacier pricing, further lowering the expense for long-term retention.[14] It maintains high reliability with 99.99% availability over a given year and 99.999999999% (11 9's) durability, achieved through redundant storage across multiple Availability Zones.[1] This storage class is particularly well-suited for backups, disaster recovery, and offsite data storage where data is accessed infrequently, such as one to two times per year, and retrieval within several hours is acceptable.[2] Objects in Glacier Flexible Retrieval require a restore request before access, and it integrates seamlessly with Amazon S3 lifecycle policies to automate transitions for cost optimization.[2] Each archived object incurs 40 KB of additional metadata storage—comprising 32 KB at the Glacier Flexible Retrieval rate and 8 KB at the S3 Standard rate—to support efficient retrieval processes.[2]Glacier Deep Archive
Amazon S3 Glacier Deep Archive is the lowest-cost storage class within the Amazon S3 Glacier family, optimized for the long-term retention of data that is accessed infrequently, such as backups and archival records requiring durable, secure storage over extended periods.[1] This class achieves its cost efficiency through optimized data placement across multiple AWS Availability Zones, ensuring 99.999999999% (11 9's) durability while prioritizing minimal retrieval needs.[1] It serves as an ideal solution for organizations seeking to store petabytes of data without the operational overhead of physical infrastructure. Retrieval from Glacier Deep Archive is available via two options: standard retrieval, which restores data within 12 hours, or bulk retrieval, which completes within 48 hours; unlike other Glacier classes, expedited retrieval is not supported to maintain the lowest pricing tier.[1] Objects in this class have a minimum storage duration of 180 days, after which early deletion incurs pro-rated charges for the remaining period.[2] Additionally, each archived object requires 40 KB of metadata storage—comprising 32 KB at the Deep Archive rate and 8 KB at the S3 Standard rate. This fixed metadata charge makes storing very small objects (under 40 KB) less cost-efficient per GB.[7] As of 2025, storage in Glacier Deep Archive costs $0.00099 per GB-month (or $1 per TB-month), representing the most affordable cloud object storage option and up to 75% lower than the S3 Glacier Flexible Retrieval class.[8][1] Launched on March 27, 2019, this storage class was introduced to deliver competitive pricing against off-premises tape archival services, enabling customers to eliminate the capital and maintenance expenses of on-premises tape libraries while providing millisecond access to metadata for efficient inventory management.[10][16] Glacier Deep Archive is particularly well-suited for compliance-driven archives in sectors like financial services and healthcare, where long-term regulatory retention is mandatory; digital preservation efforts in media and entertainment, such as storing historical footage; and as a scalable replacement for traditional tape-based systems, reducing total ownership costs through automated management and global redundancy.[1][17]Retrieval and Management
Retrieval Options
For the S3 Glacier Flexible Retrieval and S3 Glacier Deep Archive storage classes, Amazon S3 offers up to three retrieval tiers to access archived data: Expedited, Standard, and Bulk (with Expedited available only for Flexible Retrieval, and Standard and Bulk for both classes). Expedited retrieval provides the quickest turnaround for urgent access needs and is available exclusively for objects in the S3 Glacier Flexible Retrieval storage class, with provisioned capacity options to ensure dedicated throughput.[15] Standard retrieval serves as the default balanced option suitable for most use cases, while Bulk retrieval optimizes for cost efficiency when restoring large volumes of data, often at no additional retrieval fee for objects in S3 Glacier Flexible Retrieval.[15] Additionally, the AWS Free Tier includes up to 10 GB of free data retrieval per month across eligible Glacier storage classes, enabling low-volume access without charges.[18] Retrieval jobs are initiated programmatically using the Amazon S3RestoreObject API operation, where users specify the desired tier (Expedited, Standard, or Bulk) along with the object's key and restoration period. Once initiated, subsequent access occurs via GetObject for full retrieval or GetObject with range parameters for partial downloads of the restored object.[19] For legacy Amazon Glacier vaults, the InitiateJob API handles similar retrieval requests, supporting inventory, archive, or retrieval jobs.
Upon job initiation, Amazon S3 assigns a unique Job ID for tracking progress. Users monitor status by polling the HeadObject API to check restoration completion or, for vault-based operations, the DescribeJob or ListJobs APIs.[19] Notifications for job completion can be configured via Amazon S3 Event Notifications, which trigger on restore events and integrate with Amazon SNS or AWS Lambda for automated workflows.[20] In legacy Glacier setups, vault notifications publish to SNS topics upon job finalization.[21]
Provisioned capacity enhances reliability for Expedited retrievals in S3 Glacier Flexible Retrieval by reserving resources; each unit of capacity supports up to three retrievals every five minutes and a throughput of 150 MB per second, purchasable via the Amazon S3 console, API, or AWS CLI.[15] This feature mitigates potential throttling during high-demand periods without altering underlying retrieval times, which vary by storage class.[2]
Amazon S3 Select integrates with Glacier by allowing SQL-based in-place queries on restored objects in active storage tiers, enabling selective data extraction without downloading entire archives; however, direct queries on unretrieved Glacier objects require prior restoration, as standard S3 Select does not support archive tiers natively.[22] (Note: The former S3 Glacier Select feature for querying archives without restoration is available only to existing customers, with new access closed since July 2024.)[23]
Key limitations include the lack of real-time streaming access to archived objects, necessitating a restore job before any Get or Select operations.[19] Partial retrievals via S3 Select incur charges proportional to the scanned data volume, measured in GB, regardless of the returned results size.[22] All retrievals are subject to API request quotas, such as 1,000 transactions per second per prefix.[15]
Lifecycle Policies and Vaults
Amazon S3 Lifecycle policies enable automated management of object storage by defining rules that transition objects between storage classes based on age, thereby optimizing costs for archival data. These rules allow users to specify transitions from higher-cost classes like S3 Standard to S3 Glacier classes—for example, to S3 Glacier Instant Retrieval after 90 days (with a 90-day minimum storage duration in the class), to S3 Glacier Flexible Retrieval after 30 days (with a 90-day minimum storage duration in the class), or to S3 Glacier Deep Archive after 30 days (with a 180-day minimum storage duration)—subject to the target class's minimum storage duration rather than a fixed prerequisite age for the transition. For instance, a common configuration transitions infrequently accessed objects from S3 Standard-IA to S3 Glacier Flexible Retrieval after 90 days to balance accessibility and cost savings.[24][2] Vaults in Amazon Glacier represent a legacy organizational structure for the standalone Glacier service, separate from S3-integrated storage classes, where archives are stored without direct S3 bucket integration. Users create vaults to group archives logically, and each vault can hold unlimited data with access controlled via IAM policies. Vault Lock policies, introduced in 2015, provide write-once-read-many (WORM) compliance by allowing users to apply immutable retention rules using IAM policy syntax; once initiated and completed after a 24-hour validation period, the lock cannot be altered or removed. These policies support retention periods up to 100 years to meet regulatory requirements like SEC Rule 17a-4.[25][26] Vault inventory jobs facilitate content management by generating point-in-time listings of all archives within a vault, including metadata such as archive IDs, sizes, and creation dates. Initiated via the Initiate Job API, these jobs produce output in JSON (default) or CSV format, with completion typically taking up to a few hours; inventories are updated daily but reflect a snapshot from the job initiation time. The retrieved inventory can be downloaded directly or, for analysis, migrated to an S3 bucket using AWS solutions guidance to enable querying and further processing.[27][28] Expiration rules within S3 Lifecycle configurations automate the permanent deletion of objects after defined retention periods, helping control storage growth and costs in Glacier classes. For example, objects can be set to expire 3,650 days after creation following a transition to S3 Glacier Flexible Retrieval, but early expiration incurs minimum storage duration charges—90 days for S3 Glacier Instant Retrieval or Flexible Retrieval, and 180 days for S3 Glacier Deep Archive. These rules apply only to current object versions in versioned buckets, with noncurrent versions handled separately via dedicated actions.[29][30] Best practices for implementing these policies emphasize combining object tags with lifecycle rules for granular control, reducing the number of rules needed across diverse datasets. Tags, such as "TransitionArchive:90" indicating a 90-day shift to S3 Glacier Instant Retrieval, allow filtering rules to apply specific actions to subsets of objects regardless of prefix, enabling flexible management like separate retention for compliance-tagged data. Automate tag application during uploads via the PutObject API or Lambda triggers, and for existing objects, use S3 Batch Operations with inventory reports to tag at scale without incurring extra costs.[31][32]Pricing
As of November 2025 (US East - N. Virginia), Amazon S3 Glacier offers tiered storage pricing based on the selected storage class, designed for long-term archival with costs measured per gigabyte per month (GB-month). The S3 Glacier Instant Retrieval class charges $0.004 per GB-month, providing millisecond access for data that may need quick retrieval.[8] The S3 Glacier Flexible Retrieval class is priced at $0.0036 per GB-month, suitable for infrequently accessed data with retrieval times ranging from minutes to hours.[8] The lowest-cost option, S3 Glacier Deep Archive, costs $0.00099 per GB-month for data expected to remain untouched for extended periods, with retrieval times of 12 hours or more.[8] All Glacier classes apply minimum object size charges to ensure cost efficiency for small files. For Instant Retrieval, objects smaller than 128 KB are charged as 128 KB.[8] Flexible Retrieval and Deep Archive include a 40 KB minimum, comprising 8 KB of metadata at S3 Standard rates and 32 KB at the respective class rates.[8] To encourage long-term retention, each class enforces a minimum storage duration, with pro-rated charges applied for early deletions. Instant Retrieval and Flexible Retrieval require a 90-day minimum, while Deep Archive mandates 180 days; deleting data before these periods incurs fees for the remaining duration.[8] There are no fees for data ingress into S3 Glacier storage classes.[8]| Storage Class | Storage Cost (per GB-month) | Minimum Object Size | Minimum Duration |
|---|---|---|---|
| Glacier Instant Retrieval | $0.004 | 128 KB | 90 days |
| Glacier Flexible Retrieval | $0.0036 | 40 KB | 90 days |
| Glacier Deep Archive | $0.00099 | 40 KB | 180 days |
Retrieval and Request Fees
Amazon S3 Glacier incurs retrieval fees based on the chosen retrieval option and the amount of data retrieved, with variations between the Flexible Retrieval and Deep Archive storage classes. For Expedited retrieval (1-5 minutes) in Flexible Retrieval, users pay $10 per 1,000 requests plus $0.03 per GB retrieved. Expedited retrieval is also available for Deep Archive but may take 1-12 hours. Standard retrieval (3-5 hours for Flexible, 12 hours for Deep) costs $0.05 per 1,000 requests plus $0.01 per GB in Flexible Retrieval and $0.10 per 1,000 requests plus $0.02 per GB in Deep Archive, while Bulk retrieval (5-12 hours for Flexible, 48 hours for Deep) is free for both requests and data in Flexible Retrieval and $0.025 per 1,000 requests plus $0.0025 per GB in Deep Archive. A limited free tier allows up to 10 GB of Standard retrievals per month at no charge.[8] Request fees apply to API operations for managing and accessing data in Glacier buckets using Glacier storage classes. PUT, COPY, and POST requests cost $0.005 per 1,000 requests, while GET and SELECT requests are priced at $0.0004 per 1,000 requests. These fees support operations like uploading archives, initiating retrieval jobs, and querying data without full retrieval.[8] Data transfer out from Amazon S3 Glacier to the internet follows a tiered structure to accommodate varying usage volumes. The first 100 GB per month is free, followed by $0.09 per GB for the next 10 TB, with rates decreasing for higher volumes (e.g., $0.085 per GB for the next 40 TB). Transfers within the same AWS Region, such as to Amazon EC2, incur no additional fees.[8] For Select queries, which enable in-place querying of archived data without full retrieval, fees are based on the data scanned and returned. Standard Select queries charge $0.008 per GB scanned, with higher rates for Expedited queries (e.g., $0.03 per GB scanned plus request fees). Returned data incurs additional transfer costs if downloaded.[8] Vault-specific operations, such as initiating inventory jobs to list archives, cost $0.025 per 1,000 objects retrieved in the inventory report. These jobs help manage large-scale archives but add to operational expenses for frequent audits. Provisioned capacity units can be purchased for predictable high-volume Expedited retrievals at additional fixed monthly fees.[8]| Fee Type | Description | Pricing (US East, N. Virginia) |
|---|---|---|
| Expedited Retrieval (Flexible Retrieval) | Per 1,000 requests + per GB | $10 + $0.03/GB |
| Expedited Retrieval (Deep Archive) | Per 1,000 requests + per GB | $10 + $0.03/GB |
| Standard Retrieval (Flexible Retrieval) | Per 1,000 requests + per GB | $0.05 + $0.01/GB |
| Standard Retrieval (Deep Archive) | Per 1,000 requests + per GB | $0.10 + $0.02/GB |
| Bulk Retrieval (Flexible Retrieval) | Per 1,000 requests + per GB | Free |
| Bulk Retrieval (Deep Archive) | Per 1,000 requests + per GB | $0.025 + $0.0025/GB |
| PUT/COPY/POST Requests | Per 1,000 requests | $0.005 |
| GET/SELECT Requests | Per 1,000 requests | $0.0004 |
| Data Transfer Out (first 100 GB/month) | To internet | Free |
| Data Transfer Out (next 10 TB/month) | To internet | $0.09/GB |
| Select Queries (Standard, scanned) | Per GB scanned | $0.008 |
| Vault Inventory Jobs | Per 1,000 objects | $0.025 |
Use Cases and Integration
Archival Applications
Amazon S3 Glacier is widely utilized for long-term data retention in scenarios requiring durable, low-cost archival storage, such as legal holds, media libraries, and scientific datasets. Organizations employ Glacier's storage classes to preserve records that must be retained for extended periods due to regulatory requirements, ensuring data integrity over decades without frequent access. For instance, media companies archive vast libraries of historical content, like video footage and audio files, leveraging Glacier's high durability of 99.999999999% to safeguard irreplaceable assets. Similarly, research institutions store scientific datasets, including genomic sequences and environmental observations, for future analysis while minimizing ongoing costs.[1] In backup strategies, Glacier supports offsite replication for disaster recovery by integrating with Amazon S3 Cross-Region Replication, enabling data to be mirrored across geographic regions for redundancy. This approach facilitates tiered retrieval options—ranging from minutes in Flexible Retrieval to 12-48 hours in Deep Archive—allowing organizations to balance recovery point objectives (RPO) and recovery time objectives (RTO) based on data criticality. For example, critical backups can use faster tiers to achieve lower RTOs, while less urgent archives opt for deeper, cheaper storage to reduce overall expenses without compromising availability during recovery. Such strategies enhance resilience against site failures or ransomware, as demonstrated in solutions using AWS Storage Gateway for hybrid environments.[33][34] Glacier offers substantial cost savings for cold data, with storage costs up to 95% lower than Amazon S3 Standard, making it economical for infrequently accessed archives. This reduction stems from optimized pricing tiers, where Deep Archive charges as low as $0.00099 per GB per month compared to $0.023 for Standard, enabling organizations to scale petabyte-level storage affordably.[8] In healthcare, Glacier enables HIPAA-compliant retention of patient records and imaging data, as seen with providers like hc1.com archiving encrypted datasets for long-term access during audits or legal reviews. Financial institutions, such as Capital One, utilize Glacier for audit trails and regulatory compliance, storing transaction logs and compliance documents to meet retention mandates like SEC rules while optimizing resource allocation. These applications highlight Glacier's role in supporting industry-specific governance without excessive overhead.[1][35][36] A key challenge in archival applications is planning for retrieval delays, which can range from several minutes to days depending on the storage class, requiring workflows to incorporate asynchronous access patterns. Users must design applications with these latencies in mind, using tools like Amazon S3 Lifecycle policies to automate transitions and avoid unexpected costs from expedited restores.Integration with Amazon S3
Amazon S3 Glacier operates as a set of storage classes within the Amazon S3 ecosystem, enabling users to store objects directly in S3 buckets while designating them for archival purposes. This integration allows seamless object storage by uploading data to standard S3 buckets and specifying a Glacier storage class—such as S3 Glacier Flexible Retrieval (GLACIER), S3 Glacier Deep Archive (DEEP_ARCHIVE), or S3 Glacier Instant Retrieval (GLACIER_IR)—at the time of upload via the AWS Management Console, REST API, or other interfaces.[2] Unlike the legacy Amazon Glacier service, which required separate vault management, these classes treat archival objects as native S3 objects, supporting full S3 compatibility for metadata, versioning, and tagging. As of December 15, 2025, the standalone Amazon Glacier service no longer accepts new customers, directing users to the S3-integrated classes.[12][2] Lifecycle automation enhances this integration by enabling automatic tiering of objects from frequently accessed classes like S3 Standard to Glacier classes based on predefined access patterns and age thresholds. Users configure S3 Lifecycle policies in the console or via API to transition objects after a specified number of days, optimizing costs without manual intervention—for instance, moving data unused for 30 days to GLACIER.[37] This automation applies across an entire bucket or to specific prefixes/tags, ensuring data flows dynamically through the storage tiers.[37] Practical tools facilitate uploads and management within this integrated framework. The AWS Command Line Interface (CLI) supports direct designation of Glacier classes during operations, as in the commandaws s3 cp file.txt s3://my-bucket/ --storage-class [GLACIER](/page/Glacier), which uploads a local file to an S3 bucket in the Flexible Retrieval class.[38] Similarly, AWS SDKs provide programmatic support; for example, in Python using Boto3, the upload_file method accepts an ExtraArgs dictionary with 'StorageClass': '[GLACIER](/page/Glacier)' to store objects in the archival tier.[39]
Hybrid setups combine Glacier classes with S3 Intelligent-Tiering for adaptive storage, where objects automatically move to archival access tiers (e.g., Archive Access after 30 days of inactivity or Deep Archive Access after 180 days) based on real-time access patterns, without predefined rules.[40] This dynamic movement leverages monitoring to balance cost and availability seamlessly within S3.[40]
For users with data in legacy Amazon Glacier vaults prior to the 2012 launch of Amazon Glacier and its integration with S3, migration paths involve automated transfer to S3 buckets using AWS solutions like the Data Transfer from Amazon Glacier Vaults to Amazon S3 implementation, which restores and copies archives to a target bucket with the desired Glacier class.[4][41] This process eliminates the need for separate vault management, aligning legacy data with modern S3 features.[41]
Security Features
Encryption Methods
Amazon S3 Glacier secures data through server-side encryption options for storage at rest, applicable to its storage classes such as S3 Glacier Flexible Retrieval and S3 Glacier Deep Archive. These options include SSE-S3, which employs AWS-managed keys with AES-256 encryption and is automatically enabled as the default for all new objects uploaded to Amazon S3 buckets, incurring no additional cost. SSE-KMS utilizes customer-managed keys via AWS Key Management Service (KMS) for enhanced control, allowing integration with compliance requirements, while SSE-C permits customers to supply their own encryption keys, which Amazon S3 uses to handle the encryption and decryption processes.[42][43] Key rotation differs by method: AWS performs automatic rotation for the underlying keys in SSE-S3, whereas SSE-KMS relies on manual rotation configured through the AWS KMS console, API, or automated policies. SSE-C requires customers to manage key rotation independently, as keys are provided per request. All methods ensure data integrity and confidentiality using the robust AES-256 standard.[42] For data in transit, Amazon S3 Glacier mandates HTTPS with TLS 1.2 or later for all API operations and file transfers, preventing interception during upload, retrieval, or management activities. This applies uniformly to interactions with storage classes and ensures end-to-end protection without additional configuration.[44] In the legacy vault-based Amazon S3 Glacier service, archives are automatically encrypted at rest upon upload using AWS-managed AES-256 keys. The legacy service will stop accepting new customers on December 15, 2025, but existing vaults and data remain secure and accessible indefinitely. No customer key options are available. AWS handles key rotation automatically in this model.[45][46][12]Access Controls and Compliance
Amazon S3 Glacier, as a storage class within Amazon S3, utilizes AWS Identity and Access Management (IAM) policies to manage permissions at both the bucket and object levels. These policies are defined in JSON format and allow administrators to specify allow or deny actions, such ass3:GetObject for retrieving data or s3:RestoreObject for initiating retrieval jobs from Glacier storage. Conditions can be included to enforce restrictions, including source IP address ranges via the aws:SourceIp condition key, multi-factor authentication requirements with aws:MultiFactorAuthPresent, or time-based access using aws:CurrentTime.[47][48]
Bucket policies complement IAM by providing resource-based access control directly on S3 buckets containing Glacier objects. These policies can grant or deny specific actions, for example, allowing s3:GetObject on objects in the GLACIER storage class while denying deletions with s3:DeleteObject. Like IAM policies, they support conditions such as IP-based restrictions to limit access from particular networks, ensuring granular control over archival data. Bucket policies are particularly useful for cross-account access, where permissions can be delegated to external AWS accounts without modifying user-level IAM roles.[49][50]
For regulatory compliance, Amazon S3 Glacier supports standards including GDPR, HIPAA, and PCI DSS through features that enable data immutability and auditing. Vault locks, available in the legacy Amazon Glacier service, provide write-once-read-many (WORM) protection by allowing policies to be locked against modification, supporting retention requirements for financial regulations like SEC Rule 17a-4. In the current S3 Glacier storage classes, this is achieved via S3 Object Lock, which enforces WORM semantics in two modes: Governance mode, where authorized users can bypass retention with specific permissions, and Compliance mode, which prevents all modifications or deletions regardless of privileges. Object Lock supports retention periods from 1 day to a maximum of 100 years, or indefinitely, and includes legal hold functionality to indefinitely protect objects without a predefined expiration. These mechanisms help meet immutability needs for compliance frameworks by preventing premature data alteration.[1][51][52]
Auditing in Amazon S3 Glacier is facilitated through integration with AWS CloudTrail, which logs all API calls related to storage actions, including uploads to Glacier classes, retrieval initiations, and policy changes. CloudTrail captures detailed event data, such as the principal identity, request parameters, and response elements, enabling comprehensive audit trails for compliance verification. Logs can be stored durably in S3 for long-term retention and analyzed using Amazon Athena or other tools to detect unauthorized access attempts. This integration ensures traceability of all interactions with Glacier data, supporting requirements for audit logs in regulations like HIPAA and PCI DSS.[1][53]