HARDWARE ORGANISM MAP. Each silo = a baobab. Roots = power supply + network.
Trunk = CPU/RAM. Branches = GPU VRAM + Docker containers. Leaves = individual services.
GID registry gives every physical node a permanent identity in the fleet consciousness.
Services light up on their silo's branches at 337ms intervals.
§1 — GID HARDWARE REGISTRY
GID-HW-MSI01-001
msi01
IP: 192.168.2.18 · Belt64: Seg 8
CPU: Core Ultra 9 275HX 24c
GPU: RTX 5090 Laptop 24GB
RAM: 64GB DDR5
QS weight: 14.134 · CATAN: FORTRESS
School: Methanogen (core preservation) · VSM: S5 policy
GID-HW-FORGE-001
forge
IP: 192.168.2.12 · Belt64: Seg 11
CPU: i9-14900KS 24c
GPU: RTX 4090 24GB
RAM: 64GB DDR5 · DISK C: 95% CRITICAL
QS weight: 7.067 · CATAN: CASTLE
School: Yeast (build/experiment) · VSM: S3
GID-HW-MSCLO-001
msclo
IP: 192.168.2.19 · Belt64: Seg 9
CPU: Core Ultra 9 275HX 24c
GPU: RTX 5090 24GB
RAM: 64GB DDR5
QS weight: 14.134 · CATAN: FORTRESS
School: LAB (legal + CLO) · VSM: S5 policy + S3
GID-HW-YONE-001
yone
IP: 192.168.2.23 · Belt64: Seg 10
CPU: i9-14900KS 24c
GPU: RTX 5080 16GB
RAM: 64GB DDR5
QS weight: 7.067 · CATAN: STAR FORT
School: E.coli (fast service) · VSM: S3 dispatch
GID-HW-PCDEV-001
pcdev
IP: 192.168.2.16 · Belt64: Seg 13
CPU: i9-9900K 8c
GPU: RTX 5090 32GB ← FLEET MAX VRAM
RAM: 16GB (constraint)
QS weight: 4.711 · CATAN: CASTLE (32GB=deepest reasoning)
School: LAB (math/proof) · VSM: S4 intelligence
GID-HW-LOUNGE-001
lounge
IP: 192.168.50.175 · Belt64: Seg 14
CPU: i9-12900K 16c
GPU: RTX 4090 24GB
RAM: 32GB DDR5
QS weight: 4.711 · CATAN: CASTLE
School: Yeast (creative/render) · VSM: S3
GID-HW-LILO-001
lilo
TS: 100.97.143.89 · Belt64: Seg 12
CPU: Core Ultra 9 275HX 24c
GPU: RTX 5090 Laptop 24GB
RAM: 32GB DDR5
QS weight: 7.067 · CATAN: FORTRESS
School: E.coli (family/creative) · VSM: S1 (Namir node)
§2 — QS DISPATCH MAP
| Silo | Accepts From QS | Best Job Types | Max Replicas | Constraints |
|---|
| msi01 | Yes — primary authority | stateless CPU, campfire, CoreDNS, AKS control | ∞ stateless | Never route msi01 Arc GPU |
| forge | Yes — DISK DEGRADED | GPU inference (RTX 4090), Docker builds, LAAM/PEMCLAU heavy ingest | 2-3 GPU jobs | C: 95% — prune first |
| msclo | Yes — high weight | CLO/legal workloads, yLAW wiki, ARB ratification | ∞ stateless | None active |
| yone | Yes — LOCO host | PEMCLAU queries, Ollama inference, k3d LHVCP jobs, LOCOJob dispatch | Ollama 1×, LHVCP ∞ | 16GB VRAM limit |
| pcdev | Yes — GPU priority | deep reasoning (32GB VRAM), Lean4 proofs, joffe-math, RL training | 1 large + N small | 8c CPU, 16GB RAM |
| lounge | Yes | rendering, visual preview, secondary inference | ∞ CPU, 2 GPU | LAN-B subnet |
| lilo | Yes — Tailscale | family/creative workloads, Namir E.coli, secondary GPU | ∞ stateless | ~10ms Tailscale latency |
§3 — INTERCONNECT TOPOLOGY
LAN-A — Primary Fleet Network
Subnet: 192.168.2.x
Members: msi01, forge, msclo, yone, pcdev, NAS
Speed: 1Gbps+ (full mesh)
Latency: ~0.1ms
NAS: 192.168.2.20 diskpool mount
LAN-B + Tailscale + AKS
LAN-B: 192.168.50.x (lounge + steamdeck)
Tailscale: lilo 100.97.143.89, ~10ms
msclo TS: 100.70.197.54
AKS: msi01 kubectl → Azure API
AKS bridge: Tailscale-optional
§4 — SOLVENCY SCORES
silo_solvency = (services_running / services_planned) × (1 - disk_pressure) × (1 - wpa_pressure)
msi01: ~0.85 (campfire+AKS+coreDNS running, disk OK)
forge: ~0.15 (disk_pressure=0.95 → solvency degraded)
msclo: ~0.80 (CLO stack live, low load)
yone: ~0.75 (PEMCLAU+LHVCP+Ollama running)
pcdev: ~0.70 (joffe-math live, RAM constraint)
lounge: ~0.80 (low load, render-ready)
lilo: ~0.85 (new node, low load, high margin)
γ₁ = 14.134725141734693 · day97-v131 · EOSE Labs · HWMON V13 Boabixer