# Infra Documentation

> This is the documentation for a personal homelab infrastructure built around
> Kubernetes on bare metal. The clusters run [Talos Linux](https://www.talos.dev/)
> on custom-built compute nodes and [NVIDIA DGX Spark](https://www.nvidia.com/en-us/autonomous-machines/dgx-spark/)
> GPU nodes, with [Flux](https://fluxcd.io/) for GitOps delivery and
> [Ceph](https://ceph.io/) for distributed storage. Everything is declared in a
> single Git repository — Kubernetes manifests, Talos machine configs, and the
> docs you're reading now.

## Sites

- [Fairy Site](https://infra-docs.nicole.dev/sites/fairy/): Overview of the fairy site — primary homelab with compute rack, GPU nodes, and home network infrastructure

## Compute

- [DGX Spark (GB10)](https://infra-docs.nicole.dev/compute/dgx-spark/): NVIDIA DGX Spark (GB10) configuration on Talos Linux — driver selection, GPU operator, and network interfaces
- [Freckle Node](https://infra-docs.nicole.dev/compute/freckle-node/): Freckle Node — custom-built compute node purpose-built for homelab Kubernetes clusters

## Networking

- [Fairy Network Topology](https://infra-docs.nicole.dev/networking/fairy-network/): Physical network topology, IP addressing, and network segmentation for the fairy site

## Storage

- [Ceph RBD benchmarking](https://infra-docs.nicole.dev/storage/ceph-benchmarking/): How to benchmark Ceph RBD performance from a Kubernetes client with fio, what numbers to expect from this cluster, and the common gotchas that make naive runs misleading
- [TrueNAS Mini-R](https://infra-docs.nicole.dev/storage/truenas-mini-r/): iXsystems TrueNAS Mini-R — 2U rackmount NAS with 12 hot-swap bays, dual 10GbE, and IPMI

## Runbooks

- [Firmware Upgrades on DGX Spark (Talos Linux)](https://infra-docs.nicole.dev/runbooks/firmware-upgrades/): Update DGX Spark firmware via fwupd in a privileged Kubernetes pod on Talos Linux

## Guides

- [AT&T 802.1X Bypass on Firewalla Gold Pro](https://infra-docs.nicole.dev/guides/att-802.1x-bypass/): Bypass the AT&T residential gateway by running wpa_supplicant directly on the Firewalla Gold Pro

## Reference

- [Glossary](https://infra-docs.nicole.dev/reference/glossary/): Glossary of acronyms and terms used throughout the infrastructure documentation
- [Device Inventory](https://infra-docs.nicole.dev/reference/inventory/): Flat reference table of every device across all sites
- [Naming Conventions](https://infra-docs.nicole.dev/reference/naming-conventions/): Device and cluster naming conventions used across all sites