Use cases · 2026-07-20 · 8 min read
Archives are the strangest thing in a storage budget. They are the data you touch least and worry about most: finished projects, compliance records, raw footage, research outputs, the backups of backups. You will probably never read most of it. You also cannot delete it, and the day you need one file out of it, you will need it badly.
That shape — write once, read almost never, keep for years — is what cold storage tiers were built for. It is also what makes their pricing so easy to misread, because the headline number is the one part of the bill that barely matters.
The four numbers that decide the real cost
- Storage rate — the advertised price per terabyte per month. Genuinely low in cold tiers, and the only number anyone quotes.
- Minimum storage duration — delete after a month from a tier with a 180-day minimum and you are still billed for the remaining five. Archives churn more than people expect.
- Retrieval cost and latency — pulling data back is charged separately, and the cheapest tiers can take hours to restore. Bulk restores are where surprise bills come from.
- Egress — the per-gigabyte charge to move data out of the provider’s network. This is the line that makes leaving expensive, which is not an accident. We wrote about that dynamic in Vendor Lock-In and the Cost of Leaving.
Put those together and the cost of a cold archive is dominated not by keeping the data, but by the two moments you interact with it — retrieving it, and one day moving it somewhere else. A tier that is nearly free to fill can be startlingly expensive to empty.
What a real archive actually requires
| Requirement | Why it matters at a ten-year horizon |
|---|---|
| Durability without your attention | You will not be checking on it. It has to survive neglect. |
| Integrity you can prove | Corruption over years is silent. You need to verify, not assume. |
| Predictable, bounded cost | A price that can be revised annually is not a ten-year plan. |
| No exit penalty | If leaving is expensive, you are not archiving — you are renting a hostage. |
| Independence from one company | Ten years is long enough for products to be discontinued. |
| Findability | An archive nobody can navigate is functionally deleted. |
How a decentralized network prices the same job
The model is different in a way that suits archives specifically. Rather than renting capacity monthly and being charged for movement, you pay a network directly for a quantity of space and a length of time, up front, and reading the data back costs nothing. There is no egress meter, because there is no provider network boundary to cross — the file is already spread across independent operators, and retrieving it is just reading from them.
| Cold tier at a cloud provider | Decentralized network | |
|---|---|---|
| Storage cost | Very low per month, billed forever | Paid up front for the duration you choose |
| Retrieval cost | Per gigabyte, sometimes tiered by speed | Free |
| Retrieval latency | Minutes to hours in the coldest tiers | Immediate |
| Minimum duration penalty | Common — often 90 or 180 days | None, you chose the duration |
| Egress on exit | The big one | None |
| Integrity guarantee | Provider SLA | Content-addressed and certified on-chain |
| Survives the vendor | No | Yes — no vendor to survive |
| You must remember to | Watch the bill and the tier rules | Renew before expiry |
That renewal point deserves emphasis rather than a footnote. Storage on Walrus Drive is bought in epochs — you choose how long, you pay for exactly that, and if you let it lapse the network stops storing it. There is no silent auto-billing keeping your archive alive in the background, which is either the best or the worst property depending on how organised you are. The dashboard tracks expiry dates for every file and warns you ahead of time, and renewing takes one click, but the responsibility genuinely sits with you.
A subscription archive survives as long as you keep paying and they keep operating. A paid-duration archive survives exactly as long as you bought, no matter what happens to anyone.
A sensible archiving routine
- Group by retention, not by project. Seven-year compliance data and “keep forever” family archives have different needs and should not share a policy.
- Encrypt anything sensitive before upload, since a public network stores exactly the bytes you give it. Here is the workflow.
- Store a manifest with the archive — what is in it, who made it, what opens it. Ten years is long enough for the software to have moved on.
- Prefer open formats. A perfectly preserved file in a dead proprietary format is a preserved inconvenience.
- Diarise the renewal, and test a restore once a year. An archive you have never read back is a hypothesis, not a backup.
If you want to see what a decade of your actual data costs rather than a worked example, the calculator on the pricing page queries the live network and shows the real figure, including the 3% service fee, before you commit to anything. For research groups and teams with retention obligations, Solutions covers the same ground at organisational scale.