An active archive is a way of storing data you rarely use so that it stays online, searchable and recallable, while it sits on low-cost storage such as tape or object storage. Users see the archived files in a namespace they already know and get them back themselves, without filing a ticket or waiting for a restore.
For research IT teams, an active archive is the middle ground between keeping everything on expensive primary storage and locking old data away where nobody can find it.
Why research teams need one
Research data grows faster than budgets. Instruments, sequencers, microscopes and simulations produce files that are used intensely for a few weeks or months, then rarely touched again. Yet that data cannot simply be deleted: funders and journals expect it to be kept, results must be reproducible, and a later project often needs to reprocess it.
Most organisations end up with one of two bad options. Either the data stays on fast primary storage, which fills up and has to be expanded again and again. Or it goes to a backup or a tape vault, where it is safe but effectively invisible: researchers do not know what is there, and getting it back means asking IT. An active archive removes that choice.
How an active archive works
An active archive combines three things.
- One namespace. Users browse archived data through the same kind of path or file system view they use every day. The file is listed even when its content lives on tape or in an object store.
- Policies that place the data. Rules by age, size, project, owner or file type decide where each file lives and when it moves. Data that cools down moves to cheaper storage automatically; nobody has to copy it by hand.
- Self-service recall. When a user opens or requests an archived file, the system brings it back from the storage tier that holds it. Recall from object storage is fast; recall from tape takes longer but needs no human in the loop.
Behind the namespace, the data can sit on several storage types at once: disk for recent data, object storage for warm data, tape for long-term copies. Many teams also keep two copies on different media or in different locations, so the archive doubles as a protection layer.
Active archive vs. backup vs. cold storage
The three are often confused, but they answer different questions.
| Active archive | Backup | Cold storage | |
|---|---|---|---|
| Purpose | Keep data you rarely use, findable and usable | Restore data after loss or corruption | Store data as cheaply as possible |
| What it holds | The primary copy of inactive data | Copies of active data at points in time | Data written once, rarely read |
| How users get data back | Self-service, from the same namespace | Restore by IT, usually from a point in time | Request and wait, often hours or days |
| Typical media | Disk, object storage/cloud, tape | Disk, object storage/cloud, tape | Tape, deep cloud tiers |
A backup protects what you are working on today. An active archive is where finished data goes to live. Most research organisations need both, which we cover in a later article.
What to look for in active archive software
If you are evaluating an active archive, a short checklist helps:
- Open formats. Data written to tape or object storage should be readable without the archive software, for example on tape in an open format such as LTFS. Archives outlive the software that wrote them.
- Metadata that survives. Ownership, permissions, timestamps and project metadata must move with the file and come back on recall.
- Policies, not scripts. Placement rules you can read, change and audit, rather than cron jobs nobody remembers.
- Verified copies. Checksums on every transfer, and a record of what moved where and when.
- Recall users can live with. A clear view of which files are online and which will take time, and a way to recall whole projects in one request.
- No lock-in on storage. The freedom to mix tape, object storage and disk from different vendors, and to change them later.
When an active archive makes sense
An active archive pays off when a large share of your data is no longer used day to day but still has to be kept and occasionally reused. Typical signs:
- Primary storage keeps filling up, and most of what is on it has not been opened in months.
- Researchers ask IT to find or restore old datasets, and each request takes time.
- Retention rules require keeping data for years, and nobody is sure where all of it is.
- Tape is already in the data centre but is used only for backup.
If two or more of these sound familiar, an active archive is worth a look.
How NODEUM Data Archiving implements an active archive
NODEUM Data Archiving builds the active archive around containers. A container is a virtual file system that users and applications see as a normal namespace. On each container you set placement policies that decide where data lives and when it moves. The storage behind a container is organised as secondary pools of type active archive, which can be tape, S3-compatible object storage or disk.
Users keep working through the container, while the placement policies move data between secondary pools in the background. Files stay listed in the namespace and can be recalled on demand, and copies written to tape use an open format so the data stays readable for the long term.
Want to know what an active archive would look like for your data? Book an archive sizing call with the NODEUM team.