Edge Computing Architecture

Help me design an edge computing architecture with these requirements:

Use Case: [IOT/CDN/GAMING/REAL-TIME ANALYTICS/AR-VR/RETAIL]
Cloud Provider: [AWS/GCP/AZURE]
Edge Locations: [NUMBER AND TYPE OF EDGE SITES]
Latency Requirements: [TARGET LATENCY IN MILLISECONDS]
Connectivity: [ALWAYS CONNECTED/INTERMITTENT/OFFLINE CAPABLE]
Data Volume at Edge: [DATA GENERATED PER EDGE LOCATION]

Design the edge architecture across these six components:

1. Edge Compute Platform Selection
- Evaluate edge compute options including AWS Wavelength, CloudFront Functions, Lambda at Edge, Azure Stack Edge, Google Distributed Cloud, and dedicated edge hardware. Compare lightweight container orchestration using K3s, MicroK8s, or ECS Anywhere for edge workloads. Assess serverless edge options for request-level processing versus persistent edge applications. Design compute sizing for edge locations based on processing requirements and power constraints. Plan for heterogeneous edge hardware management and configuration consistency. Evaluate edge-native frameworks like KubeEdge or OpenYurt for Kubernetes-based edge management. Define the workload split between edge processing and cloud processing.

2. Data Processing and Storage at Edge
- Design edge data processing pipelines for real-time analytics and filtering. Implement edge databases such as SQLite, InfluxDB, or TimescaleDB for local data storage and query. Define data reduction strategies to minimize data transmitted to the cloud. Design edge caching layers for frequently accessed data and API responses. Implement stream processing at the edge using lightweight frameworks. Plan for edge storage lifecycle management including local retention and cloud sync. Address data consistency between edge locations and the central cloud.

3. Edge-to-Cloud Synchronization
- Design data synchronization patterns including eventual consistency, conflict resolution, and priority-based sync. Implement store-and-forward mechanisms for intermittent connectivity scenarios. Configure message queuing with local buffering using edge message brokers. Plan for bandwidth optimization using delta synchronization, compression, and batching. Design conflict resolution strategies for data modified at both edge and cloud. Implement edge-to-cloud telemetry for monitoring and health reporting. Plan for bulk data transfer during maintenance windows.

4. Edge Security
- Implement device identity and authentication using certificates, TPM, or token-based auth. Design secure boot and firmware integrity verification for edge devices. Configure network security for edge locations including firewalls, VPNs, and zero-trust networking. Implement data encryption at rest and in transit at the edge. Design physical security considerations for edge hardware in untrusted locations. Plan for security patch management and vulnerability scanning at edge locations. Configure intrusion detection and anomaly monitoring for edge environments.

5. Edge Application Lifecycle Management
- Design deployment pipelines for edge application updates using OTA mechanisms. Implement canary deployments and rollback procedures for edge software. Configure fleet management for monitoring and managing hundreds or thousands of edge nodes. Design application versioning and compatibility management across edge and cloud. Plan for A/B testing and feature flagging at the edge. Implement edge application health monitoring and automatic recovery. Design configuration management for site-specific edge parameters.

6. Resilience and Offline Operation
- Design edge applications for graceful degradation during cloud connectivity loss. Implement local decision-making capabilities for time-critical operations. Plan for queue management and prioritization during extended offline periods. Design edge-local failover for critical processing capabilities. Implement self-healing mechanisms for edge infrastructure. Plan for edge site disaster recovery and replacement procedures. Test offline scenarios regularly and document recovery procedures.

For each component provide specific service and tool recommendations, architecture patterns with diagrams, capacity planning guidelines, cost estimates for edge infrastructure, and operational runbooks.