Glossary
Cloud archive storage
Cloud archive storage is the lowest-priced tier of cloud storage, intended for data kept for years and seldom read. The low capacity price is paid for with slow retrieval, a fee on every read and a minimum storage period, so the saving holds only while the data stays untouched.
What the low price gives up
Archive tiers sit on economical media such as tape or very dense disk, and the provider manages the media, redundancy and refresh cycles. Three conditions come with the price:
| Condition | How it works | Effect on cost |
|---|---|---|
| Delayed access | Data is requested, then retrieved in batches, usually within hours; faster options cost more | Urgent reads are billed at the highest rate |
| Retrieval fee | A per-gigabyte charge on every read, on top of any transfer out | Reading the archive repeatedly erases the capacity saving |
| Minimum storage period | Commonly between 90 and 365 days, depending on tier and provider | Data deleted early is billed for the remainder |
Cold data that turns out to be warm
Bills go wrong less because of the price than because of a classification. An audit, a legal hold, a regulator's request or a new analytics project can ask for a large share of the archive at once, each read is billed at retrieval rates, and the deadline often pushes the request into the expedited option. Access history is the most reliable guide: a dataset read several times in the past year is a poor archive candidate, and one never read since it was written is a good one. The same reasoning applies to cold storage generally, where the label describes how rarely data is wanted rather than how old it is.
Finding data before paying to read it
Retrieval is billed per object, so the cost of an audit depends on how precisely the needed objects can be named. A catalog held outside the archive, with names, dates, owners and retention class, turns a request for everything from one year into a request for the few thousand objects that matter. Without one, teams retrieve whole sets to discover what they contain, and pay for all of it.
Its place at the bottom of the tier stack
Archive is the lowest step in a sequence that runs from hot to cool to archive. Rules in storage lifecycle management move objects down as they age, and the same principle applies to owned systems under storage tiering. How long a record must exist is a separate question from where it is cheapest to hold, and belongs to retention policy design.
Frequently asked questions
Is cloud archive storage the same as backup?
No. An archive holds data kept for retention or reference. A backup holds restorable copies of live systems, where restore speed matters, and retrieval measured in hours is usually too slow for a recovery.
Is deleting archived data early free?
Usually not. Data removed before the minimum storage period ends is billed as if it had stayed for the full period.