Real-Time Fleet Telematics Ingestion Architecture Spec

Role: Principal Distributed Systems Architect specializing in connected fleet telemetry and IoT stream processing.

Context

• Active Vehicle Baseline: {{fleet_size}}
• Telemetry Protocol & Ingress Channel: {{telematics_protocol}}
• P99 Ingestion-to-Query Latency Target: {{latency_sla}}
• Target Cloud Environment: {{cloud_provider}}
• Spatial Resolution Standard: {{geofence_resolution}}
• Historical Cold Storage Retention: {{storage_retention_tier}}

Task

Generate a comprehensive technical architecture specification for an ultra-reliable, real-time telematics ingestion pipeline that processes high-frequency spatial pings, detects geofence boundary events, and routes normalized transit states to downstream dispatchers.

Method

1. Calculate raw ingress throughput, network bandwidth, and message broker partition requirements based on {{fleet_size}} and {{telematics_protocol}}.
2. Design the gateway termination layer including edge authentication, payload decompression, and packet validation.
3. Architect the stream processing topology detailing windowing strategies, out-of-order event handling, and spatial indexing matching {{geofence_resolution}}.
4. Define state management patterns for vehicle presence, ignition state, and real-time odometer accumulation.
5. Formulate hot-path versus cold-path data routing schemas targeting {{cloud_provider}} storage primitives with {{storage_retention_tier}} tiering.
6. Specify dead-letter queue (DLQ) topology, circuit breaker thresholds, and backpressure recovery behaviors to satisfy {{latency_sla}}.
7. Detail monitoring instrumentation covering end-to-end processing lag, ingestion drops, and spatial query index eviction rates.

Constraints

• MUST define exact serialization formats (e.g., Protobuf, Avro) and avoid schema-less raw JSON across the event backbone.
• MUST NOT exceed {{latency_sla}} for critical geofence breach alert evaluation.
• Architectural diagrams must be represented via valid Mermaid flowchart syntax.
• Security specifications MUST mandate mutual TLS (mTLS) or hardware-token signature verification for edge devices.

Output format

Provide a technical specification structured across 5 distinct sections:

1. Ingress & Broker Sizing Model (tabular format with IOPS, bandwidth, and partition counts)
2. Stream Processing Architecture (Mermaid topology diagram + component narrative)
3. Spatial Indexing & Geofence Logic (algorithmic breakdown)
4. Storage Tiering & Data Schema (data contract definitions for hot/cold storage)
5. Fault Tolerance & Observability Matrix (SLO tables and disaster recovery runbook) Total specification should not exceed 1,800 words.

Self-review

• Validate that ingress sizing calculations mathematically scale to {{fleet_size}} transmitting over {{telematics_protocol}}.
• Confirm every edge case regarding out-of-order timestamp arrival is addressed in the stream processor spec.
• Verify all cloud services mentioned are native to {{cloud_provider}}.