Files
microdao-daarion/docs/infrastructure_quick_ref.ipynb
Apple 778907cf0e docs: add NODE3 (Threadripper PRO + RTX 3090) to infrastructure
Added NODE3 - AI/ML Workstation Specification:

Hardware:
- CPU: AMD Ryzen Threadripper PRO 5975WX (32 cores / 64 threads, 3.6 GHz boost)
- RAM: 128GB DDR4
- GPU: NVIDIA GeForce RTX 3090 24GB GDDR6X
  - 10496 CUDA cores
  - CUDA 13.0, Driver 580.95.05
- Storage: Samsung SSD 990 PRO 4TB NVMe
  - Root: 100GB (27% used)
  - Available for expansion: 3.5TB

System:
- Hostname: llm80-che-1-1
- IP: 80.77.35.151:33147
- OS: Ubuntu 24.04.3 LTS (Noble Numbat)
- Container Runtime: MicroK8s + containerd
- Uptime: 24/7

Security Status:  CLEAN (verified 2026-01-09)
- No crypto miners detected
- 0 zombie processes
- CPU load: 0.17 (very low)
- GPU utilization: 0% (ready for workloads)

Services Running:
- Port 3000 - Unknown service (needs investigation)
- Port 8080 - Unknown service (needs investigation)
- Port 11434 - Ollama (localhost only)
- Port 27017/27019 - MongoDB (localhost only)
- Kubernetes API: 16443
- K8s services: 10248-10259, 25000

Recommended Use Cases:
- 🤖 Large LLM inference (Llama 70B, Qwen 72B, Mixtral 8x22B)
- 🧠 Model training and fine-tuning
- 🎨 Stable Diffusion XL image generation
- 🔬 AI/ML research and experimentation
- 🚀 Kubernetes-based AI service orchestration

Files Updated:
- INFRASTRUCTURE.md v2.4.0
- docs/infrastructure_quick_ref.ipynb v2.3.0

NODE3 is the most powerful node in the infrastructure:
- Most CPU cores: 32c/64t (vs 16c M4 Max)
- Most RAM: 128GB (vs 64GB)
- Dedicated GPU: RTX 3090 24GB VRAM
- Largest storage: 4TB NVMe (vs 2TB)

Co-Authored-By: Warp <agent@warp.dev>
2026-01-09 05:53:16 -08:00

32 KiB