Skip to content

Latest commit

 

History

History
39 lines (29 loc) · 2.45 KB

ARCHITECTURE.md

File metadata and controls

39 lines (29 loc) · 2.45 KB
Raw

Architecture

The solution is composed of two major components:

  • A microservice - exposes REST endpoints, provides a generic connector interface which can be used by broker clients to connect to a message broker, a generic data mapper, a comprehensive expression language for data mapping and the Cumulocity Microservice SDK to connect to Cumulocity. It also supports multi tenancy.

  • A frontend (plugin) - uses the exposed endpoints of the microservice to configure a broker connection & to perform graphical data mappings within the Cumumlocity IoT UI.

The architecture of the components consists of the following components:


The light blue components are part of this project which are:
  • three default connectors for..
    • MQTT client - using hivemq-mqtt-client to connect and subscribe to MQTT brokers
    • MQTT Service client - using hivemq-mqtt-client to connect to MQTT Service
    • Kafka connector - to connect to Kafka brokers
  • Data mapper - handling of received messages via connector and mapping them to a target data format for Cumulocity IoT. Also includes an expression runtime JSONata to execute expressions
  • C8Y client - implements part of the Cumulocity IoT REST API to integrate data
  • REST endpoints - custom endpoints which are used by the MQTT Frontend or can be used to add mappings programmatically
  • Mapper frontend - A plugin for Cumulocity IoT to provide an UI for MQTT Configuration & Data Mapping

Please Note: When using MQTT or any other Message Broker beside MQTT Service you need to provide this broker available yourself to use the Dynamic Data Mapper.

The mapper processes messages in both directions:

  1. INBOUND: from Message Broker to C8Y
  2. OUTBOUND: from C8Y to Message Broker

The Dynamic Data Mapper can be deployed as a multi tenant microservice which means you can deploy it once in your enterprise tenant and subscribe additional tenants using the same hardware resources. It is also implemented to support multiple broker connections at the same time. So you can combine multiple message brokers sharing the same mappings. This implies of course that all of them use the same topic structure and payload otherwise the mappings will fail.