Regulated industries
Finance, healthcare, and government operators whose compliance frameworks require data and control to stay inside an audited perimeter.
Platform/On-Premises
Your data centre. Our control plane.
Run the full Hive console inside your own infrastructure, with the same feature set and the same policy engine, under your network perimeter, your change-control process, and your schedule.
Who it's for
When cloud isn't the answer
Some fleets sit behind requirements that no shared cloud can satisfy: a residency mandate, an audit boundary, an infrastructure team that owns everything it runs. On-premises puts Hive inside that boundary.
Data-sovereign teams
Organisations bound by residency rules that no multi-tenant or even private-cloud region can fully satisfy.
Self-hosting infrastructure teams
Teams that already run their own container platform and would rather own the control plane than trust a third party with it.
Architecture
Containerised, not a black box
The on-premises control plane ships as a set of signed container images: the same services that run our cloud, deployed onto infrastructure you own and operate.
- Kubernetes or Docker Compose reference deployments
- Console, API, device gateway, and job workers as separate services
- PostgreSQL and object storage: bring your own or use the bundled defaults
- Horizontal scaling for device gateway nodes as fleet size grows
- Devices connect outbound over TLS: no inbound firewall rules required
Requirements
What your team needs to provide
A standard sizing guide covers most fleets out of the box. Larger deployments get a custom capacity plan from our solutions team during onboarding.
Compute nodes
- 3 nodes minimum for high availability
- 8 vCPU / 32GB RAM per node baseline
- Container runtime, no proprietary hypervisor required
Storage
- Persistent volumes for database and object storage
- Snapshot-capable storage class for backups
- Sized for logs, session recordings, and APK archives
Egress rules
- Outbound HTTPS to Hive's licence and update endpoints
- Inbound TLS for device enrollment and console access
- No egress required for the fully offline variant: see air-gapped
On-premises at scale
Your perimeter, production-grade
- Nodes for high availability
- 3Nodes minimum for high availability
- Sizing plan for your fleet
- CustomCapacity plan sized to your fleet
- Data leaving your perimeter
- 0Device data leaving your perimeter
- Update cadence you control
- 100%Update cadence, on your schedule
Running it
From install to steady state
- 01 InstallationSolutions team walks your infra team through the reference deployment, credentials, and first enrolled device.
- 02 Update cadence you controlSigned release packages land on your schedule: apply them the day they ship, or batch them into your change windows.
- 03 Data stays in your perimeterDevice telemetry, session recordings, and audit logs never leave the network boundary you define.
- 04 Security responsibility splitHive ships and signs the software; your team owns host hardening, network controls, and physical security.
- 05 Backup & disaster recoveryDocumented runbooks for snapshotting, point-in-time restore, and multi-region failover. Your team owns the schedule.
| Support tier | Response window | Onboarding |
|---|---|---|
| Custom agreement | Dedicated engineering contact, contracted SLA | Custom sizing, white-glove cutover |
On-premises deployment ships under a Custom agreement, scoped with our engineers: see pricing.
Related
Keep exploring
Air-Gapped Networks
The fully offline variant, for fleets that can never touch the public internet.
Explore →Cloud Deployment
Fully managed and always current: the fastest path to your first enrolled device.
Explore →Platform Overview
One control plane, four deployment modes, and Android support from GMS to bare AOSP.
Explore →