weeve Documentation

weeve Agent & Edge Node Devices

Edge node devices run the weeve agent and edge applications
Edge node devices are hardware or virtual devices that are capable of running both the weeve Agent and the required edge applications. A detailed view is presented in the figure below.
Figure 2 - Node details
The purpose of the weeve platform is to transform data at the edge and to send only cleaned and normalized data to the desired destination (business system). The journey begins with the data points or data sources which can be either sensor, controllers or other aggregation instances that are mounted within the operating system running on the edge hardware. The first module of an edge application is typically the module that reads the raw data coming from the mounted data stream (for details on edge applications, see the relevant chapter). Further modules can process and transform the data until finally, a module sends the data to an upstream endpoint. These endpoints can be, for example, a customer database, dashboard, or messaging system.
The weeve Agent is the software process that ensures edge applications are running. To start an edge application, the weeve Agent receives a message describing the application. The Agent downloads any required container images for the modules composing the application. These images are stored on the host storage. The Agent then runs each module and establishes all connectivity required. Similarly, the weeve Agent can stop or delete edge applications.
In addition, the weeve Agent transmits status and error messages to the weeve Manager.
In order to communicate with the weeve Manager, a secure channel is established. The security depends on a public/private key pair for the weeve Agent, and a certificate for the weeve Manager.
Provisioning the edge hardware with the weeve Agent is detailed in the relevant chapter.