Azure Blob Storage Access Tiers: Hot, Cool, Cold, and Archive

Azure Blob storage offers four access tiers — Hot, Cool, Cold, and Archive — that trade storage cost against access cost and, for Archive, immediate readability. Hot costs the most to store but the least to read, Archive costs the least to store but requires a rehydration wait before you can read it, and Cool and Cold sit in between as online tiers for data touched less often. Picking the right tier for a workload’s actual access pattern is one of the most heavily tested judgment calls on the AZ-900 exam — see the full AZ-900 study guide for how this fits into all three exam domains.

The four tiers

• Hot is built for data accessed or modified frequently. It carries the highest per-gigabyte storage cost of the four tiers but the lowest cost per read or write, which is the right trade-off for data in active use.
• Cool targets data accessed infrequently, with a recommended minimum storage duration of around 30 days. It costs less to store than Hot but more per access, and — critically — it stays online, so reads are instant with no rehydration delay.
• Cold sits between Cool and Archive: an online tier, still instantly readable with no rehydration wait, for data accessed only a few times a year. It carries a lower storage cost than Cool but a longer recommended minimum duration (around 90 days).
• Archive is offline storage with the lowest storage cost of any tier, intended for data that can tolerate a retrieval delay measured in hours. A blob in Archive must be rehydrated to Hot or Cool before it can be read or modified at all.

The pattern to internalize: as you move from Hot toward Archive, storage cost drops and per-access cost rises, and only Archive actually takes data offline. Cool and Cold are both “cheaper but still instant” — the axis that separates them is how rarely the data is touched and how long you commit to keeping it there, not availability.

Comparing the tiers

Tier	Storage cost	Access cost	Availability	Typical minimum duration
Hot	Highest	Lowest	Online, instant	None
Cool	Lower	Higher than Hot	Online, instant	~30 days
Cold	Lower than Cool	Higher than Cool	Online, instant	~90 days
Archive	Lowest	Highest (plus rehydration)	Offline, hours to rehydrate	~180 days

Rehydration: getting data out of Archive

Reading an Archive blob means rehydrating it to an online tier first. Standard priority rehydration processes in the order requests are received and can take up to roughly 15 hours for typical blob sizes. High priority rehydration jumps the queue and can return smaller blobs in under an hour, for an extra fee — the option to reach for when a legal or compliance request needs a specific archived document back quickly without waiting for the standard queue. Neither priority level changes the underlying storage tier itself; both are just retrieval-speed settings applied at the moment you request rehydration.

Minimum storage duration and early deletion

Cool, Cold, and Archive each carry a minimum storage duration, and moving or deleting a blob before that period elapses triggers a pro-rated early-deletion charge — the tier’s discounted storage pricing effectively gets clawed back for the days you didn’t keep the commitment. This is the exact fee a company should expect if it archives a batch of blobs and then deletes a large fraction of them after only a few weeks, well short of Archive’s roughly 180-day minimum. There’s no way around the charge by deleting instead of moving the data, and no special account-wide lock — it’s simply a pro-rated fee tied to that specific blob.

Automating tier moves with lifecycle management

Manually re-tiering blobs as they age doesn’t scale, so Azure Storage offers lifecycle management policies: rule-based automation that moves blobs between tiers, or deletes them outright, based on age or time since last modification. A typical policy might move blobs untouched for 30 days from Hot to Cool, and blobs untouched for 180 days from Cool to Archive, entirely without custom scripts or manual intervention. This is distinct from features like blob soft delete (protects against accidental deletion) or storage account failover (switches the primary region during an outage) — lifecycle management is specifically the tool for automatic, policy-driven tiering.

A frequent multi-part scenario combines all three ideas: the lowest possible cost for year-old data, near-instant retrieval on the rare occasion someone urgently needs a file, and no manual work. That combination points to a lifecycle policy that automatically moves aged blobs to Archive, paired with High-priority rehydration for the rare urgent retrieval — Cool alone doesn’t hit the cost target, and manual monthly moves fail the automation requirement even if the eventual tier is correct.

How the AZ-900 exam tests this

• Matching a described access pattern to a tier. “Frequently modified” → Hot. “Rarely touched but must stay instantly readable” → Cool or Cold, depending on how rarely and the stated minimum duration. “Can tolerate hours of delay, lowest cost matters most” → Archive.
• Trapping you with an impossible instant-read requirement on Archive. Any scenario demanding zero-delay reads eliminates Archive immediately, regardless of how attractive its storage cost looks.
• Testing the early-deletion consequence. A scenario where data is deleted or re-tiered well before a tier’s minimum duration is asking you to recognize the pro-rated early-deletion fee, not assume the move or deletion is free.
• Urgent-retrieval-from-Archive scenarios want High-priority rehydration specifically — changing the storage account type, changing redundancy, or waiting out Standard priority don’t solve an under-an-hour requirement.

These tiering trade-offs come up constantly once you’re reading real exam stems, so work through AZ-900 practice questions that walk through cost-versus-access-speed scenarios until the tier names trigger automatically. Access tiers apply specifically to Blob storage inside a Standard General-purpose v2 account — for how that fits alongside Azure Files, Queue, and Table storage and the account types that support each service, see Azure Storage Services and Account Types Explained. Tier selection also interacts with durability: pairing a Cool or Archive tier with the wrong redundancy option can leave cost-optimized data under-protected, so it’s worth treating both decisions together. Because tier and redundancy choices both show up on the monthly bill, tracking their combined cost impact is exactly what Azure Cost Management + Billing is built for.

Quick reference

• Hot: highest storage cost, lowest access cost, for frequently used data.
• Cool: lower storage cost than Hot, still instant reads, ~30-day recommended minimum duration.
• Cold: lower storage cost than Cool, still instant reads, ~90-day recommended minimum duration, for data touched a few times a year.
• Archive: lowest storage cost, offline, requires rehydration (hours) before reading, ~180-day recommended minimum duration.
• Standard rehydration priority can take up to ~15 hours; High priority returns smaller blobs in under an hour for a fee.
• Moving or deleting data before a tier’s minimum duration triggers a pro-rated early-deletion charge.
• Lifecycle management policies automate tier transitions and deletions by blob age — no custom scripts required.
• Access tiers apply only to Blob storage in Standard GPv2 accounts, not to Premium performance-tier accounts.