Meter
Rehydration makes archived data readable; it is not delivery
Rehydration makes archived data readable in an online tier, either by changing the source tier or creating a readable copy where the provider supports that workflow. It is one stage of a restore, not delivery of the recovered bytes to their final destination.
Put your own numbers through the calculator to see what this meter does to a normal month and to the day you restore.
Rehydration makes archive data readable before final delivery
Rehydration is Azure's term for bringing an Archive blob into an online access tier. Azure supports changing the archived blob's tier or copying it to a new online blob, and it publishes Standard and High Priority rehydration options. [1]
The same recovery stage uses different vendor language elsewhere. Amazon S3 uses restore for creating temporary access to offline archive objects. Rehydration or restore makes data readable; later GET operations and egress move the bytes to the recovery destination. [2] [3]
| Step | What happens | Possible bill line | Source ref |
|---|---|---|---|
| Request archive access | A restore or rehydration operation is submitted [1] [2] | Restore or rehydration request | [1] [2] |
| Make data readable | The source changes to an online tier or a readable copy is created, depending on the workflow [1] [2] | Retrieved bytes and possible temporary or online-copy storage | [1] [2] |
| Read the object | The recovery tool issues GET or equivalent read operations [2] | Read requests | [2] [3] |
| Deliver the bytes | Readable data moves to the named recovery destination [3] | Destination-aware egress | [3] |
Rehydration language and copy behavior vary by vendor
The provider dataset has no universal rehydration field, so the table reports only workflows supported by the checked official sources. An absent named mode remains not published or does not apply to the captured class instead of being inferred from an Archive label. [4]
Azure and Amazon publish the clearest named workflows. Tigris and Vultr document Standard-copy behavior for archive restores. OVHcloud states that Cold Archive retrieval fees apply but leaves the captured retrieval rate and latency not published. [1] [2] [12] [11] [10]
Rehydration language and copy behavior vary by vendor9 rows
| Provider or class | Vendor term or workflow | Source-object treatment | Unresolved boundary | Source ref |
|---|---|---|---|---|
| Azure Blob Archive | Rehydration by changing the access tier or copying to an online blob; Standard and High Priority modes [1] | In-place tier change modifies the source tier; copy rehydration leaves the archived source and creates another blob | Exact price is region-specific and must come from the selected Retail Prices API record [5] | [1] [5] |
| Amazon S3 Glacier Flexible Retrieval and Deep Archive | Restore request creates temporary access for a selected number of days [2] | The archived source remains and a temporary readable copy is billed under the applicable online storage rule [2] [3] | Retrieval mode, object count, copy duration, and destination remain required inputs | [2] [3] |
| Google Cloud Nearline, Coldline, and Archive | No named rehydration mode in providers.json [4] | Reads use the class retrieval and request rules in the captured pricing record | A missing rehydration mode is not evidence of a temporary-copy charge | [4] [6] |
| Cloudflare R2 Infrequent Access | No named rehydration mode; reads carry retrieval and operation meters [7] | No temporary rehydrated-copy rule is published in the captured pricing evidence | Temporary-copy storage does not apply to the captured workflow, not an inferred universal promise | [7] [4] |
| DigitalOcean Spaces Cold Storage | No named rehydration mode is captured [4] | The canonical record prices retrieval but the source does not publish restored-copy behavior or recovery latency | unknown workflow details; do not invent a temporary-copy line or completion time | [4] [8] [9] |
| OVHcloud Cold Archive | Retrieval fees apply [10] | Exact readable-copy treatment is not published in the captured record | Retrieval price and latency are not published | [10] [4] |
| Vultr Archive | Archived objects must be restored and restored copies use Standard storage [11] | The archive source and Standard restored copy have separate storage treatment | Per-GB retrieval price and latency are not published | [11] [4] |
| Tigris Archive | A restore creates temporary Standard storage [12] | The Archive source remains distinct from the temporary Standard copy | Restore latency is not published in the captured pricing evidence | [12] [4] |
| Captured providers without an archive-rehydration workflow | Backblaze B2, Wasabi, IDrive e2, Hetzner, Akamai, captured Storj Standard, Filebase, and Telnyx have no named rehydration workflow in their captured primary offers [4] | does not apply to the captured class set | This is not a claim about future products | [4] |
Rehydration is read-only unless the workflow separately changes or removes the source
Copy rehydration and temporary-restore workflows preserve the archived source. Their possible bill lines are retrieval, restore and read requests, temporary or online-copy storage, and egress. They do not trigger early deletion on the unchanged source. [1] [2] [3]
An in-place tier change is different because it transitions the source object. Whether a minimum-duration remainder applies then depends on the provider, class, object age, and transition rule. That mutation must not be attached to every restore. [1] [3]
| Recovery path | Early-deletion treatment | Egress treatment | Source ref |
|---|---|---|---|
| Copy or temporary restore while preserving the source | does not apply to the read-only source [1] [2] | Apply the destination-aware transfer rule after data is readable | [1] [2] |
| In-place transition from archive to an online tier | conditional on class, object age, and the provider's transition rule [1] | Egress remains separate from the transition | [1] |
| Backblaze B2 final delivery | does not apply for a read-only restore | Free up to 3 times average monthly storage, then $0.01/GB [13] | [13] |
| Cloudflare R2 final delivery | does not apply for a read-only restore | Internet egress not charged; Class A and B operations and Infrequent Access retrieval remain separate [7] | [7] |
| Wasabi final delivery | does not apply for a read-only restore | Inside-policy egress not charged; excess per-GB pricing not published [14] | [14] |
Sources
Every figure above comes from one of these pages, on the date shown. Each numbered reference in the text links to its entry here. Vendors change prices; if a date looks old, check the source. Three vendors publish a machine-readable feed and the rest are re-checked by hand, which is why the dates are not uniform.
The source ledger14 sources
- 1Microsoft Learn: Archive rehydrate overviewlearn.microsoft.com · checked 2026-07-26
- 2AWS documentation: Restoring objectsdocs.aws.amazon.com · checked 2026-07-26
- 3Amazon Web Services: S3 pricingaws.amazon.com · checked 2026-07-26
- 4This repository: providers.jsonrepository · checked 2026-07-25T15:34:44Z
- 5Azure Retail Prices API: Retail pricesprices.azure.com · checked 2026-07-25T15:34:44Z
- 6Google Cloud: Storage pricingcloud.google.com · checked 2026-07-26
- 7Cloudflare developer docs: R2 pricingdevelopers.cloudflare.com · checked 2026-07-25T15:34:44Z
- 8DigitalOcean: Spaces object storagewww.digitalocean.com · checked 2026-07-26
- 9DigitalOcean documentation: Pricingdocs.digitalocean.com · checked 2026-07-26
- 10OVHcloud: Cold archivewww.ovhcloud.com · checked 2026-07-25T15:34:44Z
- 11Vultr documentation: Limitsdocs.vultr.com · checked 2026-07-26
- 12Tigris: Pricingwww.tigrisdata.com · checked 2026-07-25T15:34:44Z
- 13Backblaze: Transaction pricingwww.backblaze.com · checked 2026-07-25T15:34:44Z
- 14Wasabi: Faqwasabi.com · checked 2026-07-25T15:34:44Z