Comparison between platforms
Now that we have gone through the various features that an IoT platform should have, we are going to compare the five platforms and provide a matrix that can help define a use case:
Feature | AWS | Azure | Google IoT | IBM Watson | Kaa |
Device Inventory Management | Yes | Yes | Yes | Yes | No (management is more on an application level) |
Device SDK | Yes-provided | Yes-provided | Yes-can use any MQTT client software | Yes-provided | Yes-provided |
Security | Yes | Yes | Yes | Yes | Yes |
Data protocols | MQTT(s), HTTP(s) | MQTT(s) | MQTT(s) | MQTT(s), HTTP(s) | HTTP(s) |
Storage support | Yes | Yes | Yes | Yes | Yes |
Dashboard | Not built in | Not built in | Not built in | Not built in | Not built in |
Has dashboard tools as part of the platform | Yes | Yes | Yes | Yes | No |
Big data support | Yes | Yes | Yes | Yes | Depends on the infrastructure |
Integration with machine learning tools on the platform | Yes | Yes | Yes | Yes | No |
Scaling on demand | Yes | Yes | Yes | Yes | Platform supports this but it depends on the end infrastructure |
Pricing (IoT core service) # | Per million minutes of connection: $0.080. Per million messages (up to 1 billion messages... |