Cold data

Cold data is data that has not been read for a long period, measured by the last access date. Each organization sets its own dividing line, commonly somewhere between six months and three years, and cold data is the usual candidate for a move to less costly storage.

Key takeaways

Cold data is measured by last access date, and each organization sets its own dividing line.
Cold data left on the fastest tier is a recurring cost rather than a one-time one.
Finding it starts with an index that records last access for every file.

Why does cold data matter?

Because it costs far more than its use suggests. Cold data often sits on the fastest and most expensive storage, it is copied into every backup, and it slows down and inflates every migration. It also increases risk, since old files nobody reviews may still hold sensitive information. Moving or removing cold data is usually the fastest way to get capacity back.

What is the difference between hot, warm, and cold data?

The three terms describe how often data is read. Hot data is in active use and belongs on the fastest tier. Warm data is read occasionally. Cold data has not been touched for a long period. The dividing lines are set by each organization, and the last access date is what measures them.

What does it cost to keep cold data on the fastest tier?

The cost is the price difference between tiers multiplied by the capacity that should have moved. The premium is large: the VDURA Flash Volatility Index reports that enterprise solid-state cost 18.6 times as much per terabyte as hard disk in August 2026. That is down from a peak of 23.2 times in the first quarter of 2026, and still nearly three times the ratio of 7.0 a year earlier. So a petabyte left in the wrong place is a recurring number rather than a one-time one. The figure is specific to each estate, so it must be measured rather than assumed.

How do you find files nobody has opened in years?

Last access date is the mechanism. An index that records last access for every file across every system turns the question into a filter. The practical difficulties are that some storage systems update last access on backup reads, and that mount options can disable the attribute, so both need checking first.

Last accessed or last modified: which date defines cold data?

Last access is the better measure. The last modified date changes only when someone saves a change to a file, while the last access date changes when someone opens or reads it. A reference file or a finished video can be read every day and never changed, so its modified date makes it look cold when it is still in use. One caution: some storage systems turn off or delay last access updates to protect performance, and tools that open every file, such as backup or antivirus software, can reset the date. Check how each system records last access before you set a cold data policy, and where access dates cannot be trusted, use the last modified date as the fallback, together with attributes such as project status and owner.

What happens when cold data is needed again?

Files on a lower tier can still be read, but more slowly, and cloud archive tiers can take minutes to hours to restore a file and charge retrieval fees. Users also need to know where a file went. A catalog that keeps the record of every moved file lets people find it and request it back without searching each system. This is why recall rules belong in the policy from the start, not after the first complaint.

How does Diskover help?

Diskover records the last access date of every file across your storage systems, so cold data becomes a saved search you can report on and move to the right tier.

Explore data mobility

Book a demo