Industries

Clinical data outlives the hardware holding it.

Imaging retention is measured in decades; a conventional storage refresh is measured in three years. AiRE is engineered for a seven-to-ten-year service life, with capacity added while the system stays online.

  • Zero-downtime expansion
  • Per-data-set encryption
  • Immutable snapshots
  • EU data residency

The problem

Three pressures that do not resolve each other.

Imaging volume grows every year, and it does not shrink. Higher-resolution modalities, more studies per patient, and retention obligations that run for decades mean the archive only ever gets larger. Nobody in a clinical setting is going to authorise deleting a study to save capacity.

At the same time, patient data carries consequences that ordinary corporate data does not. A breach is a regulatory event, a reputational event and a clinical trust event at once. The controls have to be real and they have to be demonstrable.

And the budget is fixed by a process that runs on a different cycle from the technology. A three-year refresh assumption forces a capital conversation twice per decade for a system whose data has to survive far longer than that — which is how storage decisions end up being made by procurement timing rather than by clinical need.

What STORViX does about it

Match the platform's life to the data's life.

The mechanisms below are chosen for the specific shape of healthcare storage: an archive that only grows, data that must not leak, and continuity that is not negotiable.

  • Capacity that grows while the system runs

    Disk expansion units add 4U of capacity at a time without downtime. A base unit provides 32–501 TiB net; one DEU reaches 71 TiB–1,002 TiB; two reach 370 TiB–1.5 PiB. Up to 20 PiB fits in a single cabinet.

    • Upgradable with zero downtime
    • No forklift replacement
    • Up to 20 PiB per cabinet
  • A lifecycle that matches the obligation

    A seven-to-ten-year hardware lifecycle, achieved through reconfigurable disk packs, means the platform holding a decade of imaging is not itself replaced three times during that decade.

    • 7–10 years rather than 3
    • Disk packs reconfigured as demand changes
    • Lower TCO and embodied carbon
  • Encryption separated per data set

    Clinical imaging, administrative records and research data sets can carry different encryption and access settings on the same system, rather than one array-wide setting applied to everything.

    • Per-data-set encryption
    • Per-data-set access privileges
    • 2FA on administrative access
  • Recovery that survives a credentialled attacker

    Healthcare has been a repeated ransomware target precisely because downtime has clinical consequences. Immutable copy-on-write snapshots cannot be altered by an attacker holding administrative credentials.

    • Immutable snapshots
    • Instant revert to a known-good point
    • One-way replication into Vault
  • Integrity verified on every read

    End-to-end checksums to the root node, with automatic repair on mismatch. For a diagnostic image retrieved years after acquisition, silent corruption returning as valid data is the failure that matters.

    • End-to-end checksums
    • Self-repairing file system
    • RAID-Z or mirror topologies
  • Consolidation without compromising PACS

    Unified block and file in one unit, with per-workload profiles, so the imaging archive, the virtualisation estate and file services can share a platform without sharing one compromise setting.

    • True unified block and file
    • Performance profile for latency-sensitive work
    • Optimisation profile for the archive

Outcomes

What changes for the infrastructure team

  • Capital conversations twice a decade, not four times

    A seven-to-ten-year lifecycle removes an entire refresh cycle from the planning horizon, and with it the migration project attached to it.

  • Growth stops being an outage conversation

    Adding capacity without downtime means archive expansion is not constrained by clinical change-freeze windows.

  • Evidence at the granularity an auditor asks for

    Because encryption and access are per data set, questions about a specific category of clinical data have a specific answer.

  • A recovery path that has been rehearsed

    Instant, space-free clones make testing a restore cheap enough to do regularly rather than as an annual exercise.

Proof

The numbers behind the claim

  • 7 years

    Continuous operation without a compromising software failure, reported by INS

    STORViX customer testimonial

  • 20 PiB

    Usable capacity in a single datacentre cabinet

    STORViX published specification

  • 1.5 PiB

    Net capacity in 12U, with two disk expansion units

    STORViX published specification

  • 7–10 years

    Hardware lifecycle, against a typical three-year refresh

    STORViX published specification

Notes for a healthcare procurement process

Public healthcare procurement frequently requires a total cost comparison over the full service life rather than the purchase price. The lifecycle figure is the single largest variable in that calculation, and it is worth modelling explicitly against the incumbent's refresh assumption.

Where a tender asks for data residency, Vault replicas are held in STORViX-operated data centres in Sweden, Italy and other EU countries under a zero-knowledge privacy policy and a data processing agreement that goes beyond the statutory minimum. STORViX AB is a Swedish company, registration number 559074-0865.

Nothing on this page should be read as a claim that deploying AiRE makes an organisation compliant with any healthcare regime. The technical controls are stated so that they can be assessed; the assessment belongs to the organisation and its advisers.

Model it against your imaging growth curve.

Bring your current archive size, your annual growth rate and your retention obligation. The comparison that matters is cost over the full retention period, not the price of the first year.