Edge Device
Run Wizata on-premises, next to the machines it is reading.
An edge device is Wizata running on your own hardware, on the same site as the equipment. It is configured and deployed from the cloud platform, and it keeps working when the link to the cloud does not.
That matters for two reasons, which are the two things an edge device does:
- It connects local data sources — OPC-UA, AMQP, MQTT, Modbus, S7 and others — that are not reachable from the cloud, and buffers what they produce.
- It runs pipelines locally, close enough to the machine for closed-loop automation, without a round trip to the cloud for every execution.
Edge components require a professional license, and running pipelines on one is an add-on.
Where to start
| If you want to | Read |
|---|---|
| Understand what gets installed, and check your hardware and network are suitable | Edge Architecture & Requirements |
| Add a device to the platform and install the Wizata modules on it | Register and setup an Edge device |
| Connect a data source — a PLC, an OPC-UA server, a broker — and write values back to it | Using Edge to connect your data |
| Work out why a device is not reporting | Troubleshoot your device |
Do them in that order the first time. Registering a device that does not meet the requirements fails in ways that are much harder to read than the requirements themselves.
Running pipelines on an edge device
Once a device is registered and its modules are running, a pipeline can be deployed to it instead of to the cloud. A deployment covers several twins and decides for each one whether it runs in the cloud or on an edge device, so a single pipeline can monitor equipment in both places.
Deploying to an edge device is not the same as running in the cloud: the pipeline and its trained model are packaged into an image and delivered to the device, which then runs it on its own schedule. Training and plotting are cloud-only. The Deployment guide covers the whole sequence and what each target requires.
Updated about 1 month ago