Hot vs Cold Storage: Storage Tiers and When They Matter

Hot vs cold storage is a trade between storage cost and access cost. Hot storage costs more per gigabyte to keep but little to read, for data you use daily. Cold storage costs less to keep but more, or longer, to read back, for data you rarely touch.

TL;DR:

• Hot storage is priced for frequent reads, and cold storage is priced for data you keep but rarely open. Warm tiers sit in between.
• Azure names four tiers (hot, cool, cold, archive), and AWS splits the same range into S3 storage classes from S3 Standard to Glacier Deep Archive.
• AIOZ Storage has no storage classes in its docs. Its billing page lists one storage price, $0.017 per GB-month billed hourly, plus $0.005 per GB downloaded.

Updated Oct 19, 2026: replaced an incorrect quote about AIOZ pricing with the rates on the live billing page. We also added a tier table, a warm storage section and Azure's smart tier.

Hot vs Cold Storage: The Core Trade-Off

Every tier system is one trade with different names. Azure's documentation says the hot tier "has the highest storage costs, but the lowest access costs," and its cooler tiers flip that with lower storage costs and higher access costs. A provider can make keeping a gigabyte cheap or reading it back cheap, but the cost has to land somewhere.

That is why the right tier depends on how often you read your data. A file read thousands of times a day belongs where access is cheap, because access fees would dwarf the storage bill. A backup nobody has opened in two years belongs where storage is cheap, because instant access is a feature you aren't using.

What Is Hot Storage?

Hot storage holds data in active, frequent use. Azure calls its hot tier the best choice "for data that is in active use," and each storage account has a default tier (hot, cool or cold) for new blobs. On AWS, S3 Standard is the hot class, built for frequently accessed data with low latency and no retrieval fee.

Speed alone doesn't define hot storage, since most tiers on both clouds still return data in milliseconds, so what defines it is pricing that lets you read constantly without the bill jumping.

What Is Warm Storage?

Warm storage is the middle band, for data you still need fast but open only now and then. On Azure that is the cool tier, which expects data to stay at least 30 days. On AWS it is S3 Standard-IA, which keeps millisecond access but adds a per-GB retrieval charge. One Zone-IA costs 20% less than Standard-IA because it keeps data in a single Availability Zone.

Warm tiers make sense for monthly reports, older project files and recent backups. If you read that data more often than you planned, retrieval fees can wipe out the storage savings, so check the access logs before you move it.

What Is Cold Storage?

Cold storage holds data you keep but rarely touch. Azure's cold tier is "optimized for storing data that is rarely accessed or modified, but still requires fast retrieval," with a 90-day minimum. AWS's S3 Glacier Flexible Retrieval targets archive data "accessed 1-2 times per year," with retrieval in minutes or free bulk retrieval in 5 to 12 hours.

At the far end sits archive storage. Azure's archive tier is offline, so a blob there "can't be read or modified" until you rehydrate it, which "can take up to 15 hours." AWS calls S3 Glacier Deep Archive its "lowest-cost storage class," for data accessed "once or twice in a year" and restored within 12 hours.

How AWS and Azure Tiers Compare

The table maps each temperature band to the matching AWS storage class and Azure tier, with AIOZ Storage's single rate for contrast.

Both clouds also sell automation for teams that can't predict access. AWS S3 Intelligent-Tiering moves objects between tiers by observed use, with a 40% lower-cost infrequent tier and up to 95% savings in its deep archive tier, and no retrieval charges. Azure's newer smart tier moves data between hot, cool and cold based on usage patterns.

Azure also charges you for changing your mind early. Moving or deleting a blob before 30 days in cool, 90 in cold or 180 in archive triggers an early deletion charge for the remaining days.

The Two Costs You Trade: Retrieval Fees and Wait Time

Two costs grow as data gets colder, and you should track them separately. The first is the per-GB retrieval fee that colder tiers add on top of the storage rate, and the second is wait time, which only shows up at the archive end: hours for Glacier Deep Archive and up to 15 hours for Azure archive.

If you pick a cold tier for data you sometimes need back quickly, you pay both costs at once, right when you can least afford the wait.

How AIOZ Storage Handles It

AIOZ Storage doesn't sell storage tiers, and its docs list no storage classes. The billing page charges $0.017 per GB per month for storage, calculated every hour, and $0.005 per GB downloaded. Every object costs the same whether you read it once a year or a thousand times a day.

You give up the savings a correctly tiered archive brings for data nobody touches, and in return there are no early deletion charges, no rehydration waits and no retrieval fees beyond the flat download rate. Uploads use the standard S3 API, for example this command from the CLI docs:

export AWS_REQUEST_CHECKSUM_CALCULATION=when_required
aws configure set default.s3.addressing_style path
aws s3 --endpoint-url https://s3.aiozstorage.network cp /path/to/file s3://bucket-name

The first two lines are ours, not from AIOZ's CLI docs. AIOZ Storage rejects uploads without a Content-Length header, which AWS CLI v2.23.0 and later send by default unless you set when_required (AWS CLI environment variables). Path-style URLs (AWS CLI S3 configuration) are needed because bucket subdomains don't resolve. We confirmed both on Oct 19, 2026; a full upload with real keys is not tested yet.

The docs apply the same hot and cold logic to vector search. The vector database cost page suggests a "tiered search" setup where you keep your most-searched vectors in a fast store and the rest in AIOZ Storage. How AIOZ Storage Pricing Works has a worked cost example.

Next step: estimate your monthly downloads per terabyte stored, then compare that number against the tier fees above.

How to Decide Which Tier Fits Your Data

Start by estimating how often each dataset is read, since every tier system above prices on that one number. Data read constantly belongs in a hot tier or on a flat rate, where access is cheap. Large data read a few times a year is where a cold or archive tier on AWS or Azure saves money, as long as the retrieval wait is acceptable.

When the pattern is mixed or hard to predict, a flat rate or an automatic tiering service removes the guesswork, and you trade away some savings on the data nobody touches, but you also avoid the penalties for guessing wrong.

Frequently Asked Questions

What is the difference between hot and cold storage?
Hot storage costs more to keep but less to read, so it suits data you use often. Cold storage costs less to keep but more, or longer, to read, so it suits data you rarely open.

What are Azure's storage tiers called?
Hot, cool, cold and archive, from most to least expensive to store, and archive is offline and needs a rehydration step of up to 15 hours before you can read the data.

What are the AWS storage classes for hot and cold data?
S3 Standard is the hot class, and S3 Standard-IA and One Zone-IA cover infrequent access with a per-GB retrieval charge. Three Glacier classes cover cold and archive data, ending with Glacier Deep Archive.

Does AIOZ Storage have hot and cold storage tiers?
No storage classes appear in its docs. The billing page lists one price, $0.017 per GB-month billed hourly, plus $0.005 per GB downloaded, whatever the access pattern.

Why do cold tiers charge extra to retrieve data?
The low storage price assumes you will rarely read the data, so the retrieval fee and, at the archive end, the wait make that pricing work for the provider.

What happens if I delete data from a cold tier too early?
On Azure, moving or deleting a blob before 30 days in cool, 90 in cold or 180 in archive triggers a prorated early deletion charge for the remaining days.

Is a flat storage rate ever better than tiers?
Yes, when your access pattern is mixed or unpredictable, because you can't misfile data into the wrong tier. For a large, predictable archive that nobody reads, a cold or archive tier usually costs less.

Related Reading

• How AIOZ Storage Pricing Works - the flat-rate billing, a worked cost example and a rate comparison.
• What Is Amazon S3? Object Storage Explained - the bucket, object and key model behind every S3 storage class.
• Data Durability vs Availability: What 11 Nines Means - the other pair of guarantees AWS and Azure define.
• AIOZ Storage Explained: S3-Compatible Decentralized Storage - what AIOZ Storage is overall.
• What Is a CDN and How Does It Relate to Object Storage? - how bytes reach users once you have picked where they live.
• The 3-2-1 Backup Rule - where cold and archive copies fit in a backup plan.

References

• AWS: Amazon S3 Storage Classes
• Microsoft: Access Tiers for Blob Data (Azure Storage)
• AIOZ Storage Docs: Billing
• AIOZ Storage Docs: Vector Database Cost and Use Cases