Real time operational intelligence for multi tenant teams. Ingest metrics, monitor live dashboards, and get alerts when values breach thresholds or when forecasts and anomaly signals say trouble is coming.

One line pitch: a lightweight Datadog meets PagerDuty built on .NET 9, Angular 19, and SQL Server.

OpsDash is not only charts and static thresholds. The intelligence layer rolls statistics on every ingestion: Z score anomalies, short horizon correlation across metrics, forecasts with weighted moving average or linear regression, predictive rules that watch forecasted values, a composite tenant health score, and automatic incident timelines when anomalies cluster in time.

[Angular 19 SPA] -> [ASP.NET Core 9 Web API] -> [SQL Server]

Frontend: standalone components, Signals with RxJS, lazy loaded routes, Angular Material, Chart.js.

Backend: clean architecture, dependency injection, JWT with RBAC, middleware pipeline, FluentValidation.

Database: EF Core 9, TenantId global query filter, stored procedures, indexed views, seed data.

Real time: SignalR hub with tenant scoped groups.

Caching: IDistributedCache (in memory in development, Redis in production style configurations).

Cloud: Azure App Service, Azure SQL, Azure Cache for Redis, Azure Blob Storage, Application Insights.

Azure deployment instructions will be added in Phase 3.

The product uses a shared database model: every tenant scoped row carries a TenantId. Isolation relies on three cooperating layers: middleware that resolves the tenant from the JWT, an EF Core global query filter so queries only see the current tenant unless code explicitly opts out, and API authorization so routes cannot be abused across tenants.

• Multi tenant architecture: shared database, EF Core global query filter, tenant resolution middleware
• JWT authentication with refresh tokens
• Role based access control: SuperAdmin, Admin, User
• User management with search, pagination, and role assignment
• Metric ingestion (single and batch) via POST /api/v1/metrics
• Dashboard with KPI cards, line and bar charts, date range filter, category filter
• Threshold based alert rules, configurable per metric
• Swagger and OpenAPI with JWT auth support
• Health check endpoints: /health, /health/ready
• Server side pagination and sorting across list APIs
• Environment based configuration with user secrets for local development

• Anomaly detection: rolling Z score per metric per tenant, tiered severity (Warning, Critical, Severe), baselines recalculate on each ingestion
• Metric correlation: when an anomaly fires, a time window check (about fifteen minutes before and after) surfaces related metric movements
• Forecasting: weighted moving average (default) and linear regression; forecast appears as a dotted projection on charts
• Predictive alerting: rules can evaluate forecasted values with a configurable horizon, not only the latest raw sample
• Tenant health score: composite 0 to 100 score from normal metric share (40%), anomaly density (30%), trend direction (20%), mean time to acknowledge (10%), color coded green, yellow, red
• Incident auto grouping: anomalies inside rolling thirty minute windows cluster into incidents with lifecycle Open, Acknowledged, Investigating, Resolved, plus a full event timeline per incident

• Real time notifications: SignalR tenant scoped groups and toast style alerts in Angular
• Caching: IDistributedCache, tenant scoped cache keys, invalidation hooks tied to ingestion
• Report export: CSV and PDF artifacts stored in Azure Blob Storage
• Audit logging: EF Core SaveChanges interceptor, old and new values as JSON, admin viewer in the app

Use a trusted connection to LocalDB while developing:

Server=(localdb)\MSSQLLocalDB;Database=OpsDash;Trusted_Connection=true;TrustServerCertificate=true

Place this in user secrets or appsettings.Development.json for OpsDash.API as your project already expects.

cd src/OpsDash.API
dotnet run

Default HTTPS URL from launchSettings.json: https://localhost:7198

Swagger: https://localhost:7198/swagger

cd src/opsdash-client
ng serve

App URL: http://localhost:4200

Seed data loads three demo tenants with about ninety days of history, embedded anomalies, open incidents, and precomputed health scores.

TechFlow (SaaS) sample metrics: active_users, monthly_revenue, churn_rate, api_response_time, error_rate

Metro Health (Healthcare) sample metrics: er_wait_time, bed_occupancy, patient_throughput, staffing_ratio

UrbanCart (Ecommerce) sample metrics: checkout_conversion, cart_abandonment, page_load_time, payment_success_rate

Tables are grouped by domain. Every business table participates in multi tenancy through a TenantId column where applicable.

OpsDash/
  OpsDash.sln
  README.md
  src/
    OpsDash.API/            Controllers, middleware, Program.cs
    OpsDash.Application/    DTOs, interfaces, services, validators, mappings
    OpsDash.Domain/         Entities, enums, domain interfaces
    OpsDash.Infrastructure/ EF Core, repositories, seed, cache, tokens, tenant services
    opsdash-client/         Angular 19 standalone SPA
  tests/
    OpsDash.UnitTests/
    OpsDash.IntegrationTests/

From the repository root:

dotnet test OpsDash.sln

This runs xUnit projects under tests/OpsDash.UnitTests and tests/OpsDash.IntegrationTests.