2.5 KiB
2.5 KiB
layout, title, date, categories, highlight
| layout | title | date | categories | highlight | |
|---|---|---|---|---|---|
| post | Release 0.1: The Severed Ecosystem | 2025-12-27 02:00:00 -0400 |
|
true |
This blog serves as the public documentation for Severed. While the main site provides the high-level vision, this space is dedicated to the technical source-of-truth for the experiments, infrastructure-as-code, and proprietary tooling that are used within the cluster.
Ecosystem
The following services are currently active within the severed.ink network:
- ChronoFlow Quant: Proprietary Trading Pipeline
(Auth Required) - Flow Admin: Dagster Orchestration Engine
(Auth Required) - Optics: Centralized Grafana Observability
(Auth Required) - Gitea: Version Control & Git Ops
(Public) - Blog: Public Engineering Documentation
(Public)
Core Infrastructure
The ecosystem is powered by a Home Server Cluster managed via a Kubernetes (k3s) distribution. This setup prioritizes local sovereignty and GitOps principles.
- CI Pipeline: Automated build and test suites are orchestrated by a private Jenkins server utilizing self-hosted runners.
- GitOps & Deployment: Automated synchronization and state enforcement via ArgoCD.
- Data Layer: Persistent storage managed by PostgreSQL.
- Telemetry: Full-stack observability provided by Prometheus (metrics) and Loki (logs) via Grafana.
- Security Layer: Push/Pull GitOps operations require an active connection to a WireGuard (VPN) for remote access.
Roadmap
Engineering efforts are currently focused on the following milestones:
- OSS Strategy: Transitioning from a hybrid of AWS managed services toward a ~100% Open Source Software (OSS) stack.
- High Availability (HA): Implementing a "Cloud RAID-1" failover mechanism. In the event of home cluster instability, traffic automatically routes to a secondary cloud-instantiated Kubernetes cluster as a temporary failover.
- Data Resilience: Automating PostgreSQL backup strategies to ensure parity between the primary cluster and the cloud-based failover.
- Storage Infrastructure: Integrating a dedicated TrueNAS node to move from local SATA/NVMe reliance to a centralized, redundant storage architecture.
Terminal Redirect
For the full technical portfolio and expertise highlights, visit the main site: