Documentation
Customer Cloud
REAP deployed in your AWS (opens in a new tab), Azure (opens in a new tab) or GCP (opens in a new tab) account so that data stays in your cloud and uses your services.
Status By arrangement: scoped and scheduled per pilot with your cloud and identity teams; the timeline is agreed before week 0.
Architecture
- REAP platform components deployed into a dedicated VPC/VNet in your account (Kubernetes-based) confirm deployment method: Terraform (opens in a new tab), Helm (opens in a new tab), marketplace.
- Your KMS keys for encryption at rest; your identity provider for SSO; your PAM for device credentials.
- Connector cluster in your network reaches the platform over private connectivity or HTTPS.
Requirements
| Requirement | Typical value |
|---|---|
| Cloud account and permissions to deploy | minimum IAM policy |
| Compute | node sizes and counts for pilot scale (50–250 devices) |
| Storage | volumes and object storage for telemetry retention |
| Networking | Private connectivity from the connector cluster; egress for model endpoints customer-endpoint options |
| GCP (opens in a new tab) | GCP (opens in a new tab) as a hosting location by arrangement; GCP network discovery is not available today |
When to choose it
- When your security review requires the platform and its data in your own cloud account. Confirm availability and scope with REAP before you plan a pilot on this model.
- When your security review requires your keys and your logging.
- When you already operate cloud infrastructure and prefer to host the platform.
Engineering to confirm
- Deployment method and minimum IAM/RBAC policy per cloud
- Sizing table for pilot and production scale
- Operating responsibilities matrix (REAP vs. customer) for patching, backups and upgrades
Last reviewed: October 7, 2026 (draft) · Owner: solutions engineering · Corrections: security@reapnet.ai