Homelab Hardware Fleet Deployment
Overview
Strategic deployment guide for transforming a heterogeneous hardware fleet into a tiered, role-specialized infrastructure. Covers Mac Mini headless server conversion, Lenovo fleet revitalization (P330 Tiny, ThinkStation C30, M92p clusters), and the transition from hobbyist homelab to professional ops-core architecture.
Tiered Architecture
Always-On Core (P330 Tiny)
- Upgrade to 32GB RAM + 1TB NVMe SSD
- 8th Gen Intel for modern efficiency
- Handles critical infrastructure: reverse proxy, DNS, monitoring
Storage Vault (ThinkStation C30)
- Repurpose as centralized ZFS storage server
- ECC Registered DDR3 memory for data integrity
- High-capacity enterprise CMR HDDs for TrueNAS SCALE
Experimentation Cluster (M92p x3)
- k3s or Docker Swarm cluster for high-availability experiments
- Budget SATA SSDs sufficient for ephemeral workloads
- Isolated from production services
Build Mule (M4 Mac Mini)
- High-performance Apple Silicon for iOS development and builds
- OrbStack or Tart instead of Docker Desktop to reduce overhead
- Headless server conversion: SSH + Screen Sharing, disable sleep, automatic power-on recovery
- Separate OS from data via external USB4/Thunderbolt enclosures
Deployment Principles
- Infrastructure as Code: Every machine scriptable and recoverable from bare metal in <2 hours (Ansible/Terraform)
- VLAN segmentation: Isolate management, storage, and development traffic
- VPN-only access: Tailscale overlay network, no port forwarding
- 3-2-1 backup: ZFS snapshots + encrypted off-site replication
- Power optimization: Tiered architecture (always-on, burst compute, experimentation) balances efficiency
See Also
- iron-cloud-architecture - Sovereign compute cluster architecture and hardware role specialization
- homelab-audit-optimization - Current state audit and optimization strategies
- ssh-fleet-management - Config-driven fleet management across distributed machines