Architecture
Cloud Native Kubernetes Platform
A multi-service platform built on Azure Kubernetes Service for enterprise processing, designed for independent deployability, scalability, and team ownership.
Architecture Goals
Every architectural decision traced back to these five objectives.
Independent Deployability
Teams deploy services without coordination
Independent Scalability
Each service scales based on its own demand
Resilience
Failure in one service does not cascade
Observability
Every service exposes health, metrics, and traces
Team Ownership
Service boundaries match team boundaries
Platform Services
Nine services form the platform boundary, each owned by a dedicated team with clearly defined interfaces.
API Gateway
Central ingress for all external requests. Handles authentication, rate limiting, and request routing.
EdgeIdentity Service
Authentication and authorization across the platform. Integrates with Azure AD and manages service-to-service identity.
PlatformMetadata Service
Configuration and reference data management. Provides schema validation and versioning for all platform entities.
PlatformWorkflow Service
Orchestrates long-running business processes with state persistence, retry policies, and compensation logic.
CoreProcessing Engine
Distributed computation engine supporting fan-out, aggregation, and stream processing for enterprise workflows.
CoreReporting Service
Generates reports from processed data. Supports multiple output formats and delivery channels.
CoreNotification Service
Manages outbound communications across email, webhook, and SignalR push channels.
InfrastructureAudit Service
Immutable audit trail for all platform operations. Supports compliance and forensic analysis.
InfrastructureMonitoring Platform
Centralized observability aggregating metrics, logs, and traces from every platform service.
InfrastructureScalability Characteristics
API Layer
per API gateway instance
internal service calls
Processing Engine
messages per partition
per processing node
Workflow Services
concurrent executions
long-running workflow support
SignalR
per Azure SignalR unit
broadcast rate
Observability Pipeline
Every service emits structured telemetry through a unified pipeline. Metrics, logs, and traces converge into a single observability platform.