Deployment and device management
How do we use the platform for model deployment.
The Scailable platform allows you to (mass) deploy models to target edge devices. Effectively you can "swap" the models that run on an edge device (configured using the AI manager) remotely and change the device configuration. The latter you can do at a large scale: you can flexibly group devices and deploy models to groups of devices.
The DEVICES tab on the scailable platform shows all devices that have been registered to your account. Device registration is the first step in setting up your AI manager on your edge device: the result will be a listing of the device in this overview.
Note that here you can manage the device name, see its activity, and see additional details regarding the device installation. On this page you can also group devices for mass deployment.
Next to listing your devices, the Scailable platform allows you to deploy models to your devices. Although deployment can be done directly from the device, for mass-deployment using the Platform is easier and more scalable.
To deploy a model to a device simply select a model, and subsequently select a device or a device group (for mass deployment) to assign the model to the device.
To deploy a model to a group of devices first select the group in the device listing and then click the "Assign to {groupname}" button
In the next step you will be guided to the models lsiting where you can choose a model to assign.
At the top of the page the blue bar allows you to cancel the assignment.
All of our edge devices regularly "ping" our servers to inspect model and configuration deployments whenver they have access to our cloud. After assigning a model to a device the device will download the assigned model at the first opportunity.
On the cloud on each model detail page a listing of the devices where this model is used is displayed.
Alternatively, the used model is also shown on the detail page for each device.
Last modified 4mo ago