Cloud-native distributed monitoring platform with AI-powered schema normalisation and real-time anomaly detection.
When a data source updates and renames a field — "CPU %" becomes "processor_usage" — traditional monitoring tools silently return no data. Nobody knows why. Customers lose visibility.
A 6-tier schema normalisation engine that resolves attribute name drift automatically. No manual remapping required.
Old tools: "CPU %" → exact match → ✗ not found → NO DATA
SysMonitor: "CPU %" → fuzzy match → ✓ cpu_percent → data flows
Built across 5 independently deployable microservices on Azure AKS, with ML.NET anomaly detection and a real-time Vue 3 dashboard updated via SignalR.
C# / ASP.NET Core 8 · Vue 3 / TypeScript · Azure (AKS, Cosmos DB, Service Bus, SignalR, Key Vault) · ML.NET · Kubernetes · Bicep IaC · GitHub Actions
Prerequisites: Docker Desktop, .NET 8 SDK, Node.js 20+
git clone https://github.com/YOUR_USERNAME/SystemMonitor.git
cd SystemMonitor
start-dev.bat| URL | |
|---|---|
| http://localhost:5173 | Dashboard |
| http://localhost:5000/swagger | API |
Starts Cosmos DB + Azurite emulators, API, Worker, Agent (3 mock agents), and Dashboard automatically.