Skip to main content
Version: latest

Deployment options

VaultPAM is available in three deployment models. The right choice depends on your data sovereignty requirements, IT operations capacity, and connectivity constraints.

Comparison

FactorSaaSOn-premisesHybrid
Data locationGCP europe-central2 (Warsaw, Poland) -- data does not leave the EUYour own data centreControl-plane in your DC; connector layer in cloud or mixed
MaintenanceManaged by VaultPAM -- no patching requiredFull responsibility of your IT teamShared; cloud components managed by VaultPAM
ConnectivityOutbound-only connector; no inbound firewall changesInternal network; no external dependencyConnector phones home to control-plane
Compliance postureEU data residency, GDPR Art. 32 covered by platform DPAYou control the stack; you own the evidencePartial coverage; check with your DPO
Time to valueMinutes (sign up, install connector)Days to weeks (infrastructure provisioning)Days (hybrid onboarding path)

SaaS

VaultPAM SaaS runs on Google Cloud Platform in the europe-central2 (Warsaw, Poland) region. All tenant data -- session recordings, audit logs, credentials, and configuration -- is stored within this region. Data does not leave the EU.

This deployment satisfies GDPR data residency requirements for EU-based organisations and aligns with NIS2 Article 21 obligations without additional customer action.

Getting started with SaaS

  1. Go to app.vaultpam.com and sign up for a free trial.
  2. Complete your organisation profile (name, timezone, billing contact).
  3. Download and install the connector on a machine inside your network.
  4. Register your first resource (RDP, SSH, or web target).
  5. Launch your first session to confirm the data path is working.
EU data residency confirmation

To receive written confirmation that your data is stored exclusively in the EU, contact support@vaultpam.com to request a Data Processing Agreement (DPA). The DPA specifies GCP europe-central2 as the sole processing region.

On-premises

The on-premises model runs the entire VaultPAM control-plane on your own infrastructure. You are responsible for provisioning, patching, and backing up all components.

Prerequisites

  • Kubernetes cluster (K8s 1.28+) or Docker Compose environment
  • PostgreSQL 15+ database
  • Accessible MinIO-compatible object store (or AWS S3) for recordings
  • Outbound internet access from the control-plane host (for licence validation)

Getting started with on-premises

  1. Obtain the on-premises deployment package from support@vaultpam.com.
  2. Provision the required infrastructure (PostgreSQL, object store, K8s or Compose).
  3. Run the installer and specify on-premises mode.
  4. Set the CONTROL_PLANE_URL environment variable on each connector to point to your instance.
  5. Log in to your self-hosted instance and complete organisation setup.
  6. Install the first connector and register your first resource.
  7. Verify the audit log records your setup activity.

Hybrid

The hybrid model places the VaultPAM control-plane on-premises while allowing connectors to operate in cloud-hosted environments (or vice versa). This is common when you want to keep audit data on-premises but have workloads in AWS, Azure, or GCP.

Getting started with hybrid

  1. Deploy the control-plane on-premises following the on-premises steps above (steps 1-5).
  2. For each cloud environment, install a connector on a VM or Kubernetes pod inside that VPC.
  3. Configure each connector to reach your on-premises control-plane (VPN or private link recommended).
  4. Register cloud-hosted resources in the console.
  5. Verify connectivity by launching a test session against a cloud resource.

Which should I choose?

Quick decision guide
  • SaaS - you want the fastest time to value and EU data residency without managing infrastructure.
  • On-premises - regulations or internal policy require data to stay on your own hardware.
  • Hybrid - you have mixed environments and want a single control-plane for all of them.

When in doubt, start with SaaS. Migration to on-premises or hybrid is supported with data export tooling.