Link copied to clipboard!
Group 17: IoT & Edge Computing
Device connectivity and edge processing: IoT Core, Greengrass, SiteWise, IoT Analytics, Wavelength.
Edge Principle: Process data closer to devices for reduced latency, bandwidth efficiency, and real-time decision making.
Services & Roles
IoT Core
Managed MQTT broker with device registry and rules engine.
- Device connectivity
- Message routing
- Device shadows
Greengrass
Edge runtime extending AWS to local devices with Lambda functions.
- Local Lambda execution
- ML inference
- Device management
SiteWise
Industrial IoT data collection and analysis from equipment and processes.
- Asset modeling
- Data historians
- Equipment monitoring
IoT Analytics
Analytics service for IoT data with built-in ML and visualization tools.
- Data pipelines
- Time-series analytics
- Anomaly detection
Wavelength
Ultra-low latency applications with 5G networks at edge locations.
- 5G edge zones
- Mobile edge compute
- Ultra-low latency
Key Differences
| Dimension | IoT Core | Greengrass | SiteWise | IoT Analytics | Wavelength |
|---|---|---|---|---|---|
| Purpose | Device connectivity | Edge computing | Industrial IoT | IoT data analytics | Mobile edge |
| Location | Cloud-based | Edge devices | Industrial sites | Cloud analytics | 5G edge zones |
| Protocol | MQTT, HTTPS | Local + MQTT | OPC-UA, Modbus | Various protocols | Standard networking |
| Processing | Rules engine | Local Lambda | Asset models | Analytics pipelines | EC2 instances |
| Latency | Cloud latency | Local (ms) | Near real-time | Batch/stream | Ultra-low (sub-10ms) |
Selection Model
0–10 sliders weight IoT requirements, latency needs, and processing patterns.
Score_IoTCore = 0.32*C_deviceConnectivity + 0.26*C_mqttMessaging + 0.18*C_rulesEngine + 0.12*C_deviceRegistry + 0.08*C_deviceShadows + 0.04*C_cloudIntegration
Score_Greengrass = 0.30*C_edgeComputing + 0.26*C_localProcessing + 0.18*C_offlineCapability + 0.14*C_edgeLambda + 0.08*C_localMLInference + 0.04*C_deviceManagement
Score_SiteWise = 0.34*C_industrialIoT + 0.28*C_assetModeling + 0.16*C_equipmentMonitoring + 0.12*C_dataHistorians + 0.06*C_opcuaProtocol + 0.04*C_processOptimization
Score_IoTAnalytics = 0.32*C_iotDataAnalytics + 0.26*C_timeSeriesAnalysis + 0.18*C_anomalyDetection + 0.12*C_dataPipelines + 0.08*C_mlIntegration + 0.04*C_dataVisualization
Score_Wavelength = 0.36*C_ultraLowLatency + 0.28*C_mobileEdgeCompute + 0.16*C_fiveGIntegration + 0.10*C_edgeZones + 0.06*C_mobileApplications + 0.04*C_carrierPartnership
IoT Core {{vm.scores.iotcore|number:2}}
Greengrass {{vm.scores.greengrass|number:2}}
SiteWise {{vm.scores.siteWise|number:2}}
IoT Analytics {{vm.scores.iotanalytics|number:2}}
Wavelength {{vm.scores.wavelength|number:2}}
Primary Emphasis: {{vm.recommended.name}} ({{vm.recommended.score|number:2}})
Heuristics
- High C_deviceConnectivity + C_mqttMessaging → IoT Core for device management.
- Strong C_edgeComputing + C_localProcessing → Greengrass for edge logic.
- High C_industrialIoT + C_assetModeling → SiteWise for manufacturing.
- C_iotDataAnalytics + C_timeSeriesAnalysis → IoT Analytics for insights.
- C_ultraLowLatency + C_mobileEdgeCompute → Wavelength for mobile apps.
Anti-Patterns
- Using IoT Core for heavy local processing (use Greengrass instead).
- Using Wavelength for non-mobile edge compute (use Greengrass).
- Using SiteWise for consumer IoT (use IoT Core).
- Complex analytics without IoT Analytics (processing inefficiency).
Summary
Choose IoT Core for connectivity, Greengrass for edge, SiteWise for industrial, IoT Analytics for insights, Wavelength for mobile.