Skip to main content

Devices Overview

The Devices tab provides a centralized interface for viewing and managing all enrolled edge devices in your organization.

Device families​

Optra Edge devices come in two families, each suited to different deployment scenarios:

Compute​

Compute devices are general-purpose edge computing nodes. They run skills, process data, and communicate with the portal — including vision-based AI workloads. The Optra CX2000 includes an NPU (neural processing unit) for AI inference. Use compute devices for workloads such as network monitoring, audio processing, data aggregation, and lightweight AI inference.

Models: Optra Compute 550, Optra CX1000, Optra CX2000

Vision​

Vision devices are powered by NVIDIA GPUs, making them ideal for compute-intensive AI workloads such as object detection, image classification, and real-time video analysis.

Models: Optra VZ1000, Optra VZ5000, Optra VZ5100, Optra VZ6000, Optra VZ6100

See Device Models for a full comparison of available hardware.

Filtering devices​

The left panel contains a comprehensive set of filters to narrow down your device list:

FilterDescription
ModelFilter by device hardware model
TagsFilter by custom tags applied to devices
Firmware VersionFilter by current firmware version
WorkflowFilter by assigned workflow
SkillFilter by running skill
OutputsFilter by current skill output values

Additional filter attributes include: name, location, downtime, connection status, disk usage, memory, CPU, IP address, MAC address, audio connection, internal temperature, serial number, and enrollment date.

Searching and map view​

  • Use the Search Devices bar to locate a device by name or serial number.
  • The Device Count indicator shows the total number of devices visible under current filters.
  • Toggle Show Map to display the real-time geographical location of your devices on an interactive map, with pins color-coded by connection status:
Pin colorMeaning
🟢 GreenOnline
🟡 YellowPartial connection or updating
🔴 RedOffline
🔵 BlueUnknown / not yet connected

By default, map pin color reflects connection status. You can change this to reflect the value of a specific skill output using the dropdown above the map — useful for at-a-glance monitoring of a metric like "alert state" or "person count" across all devices.

Connection status​

Each device displays a connection status badge:

StatusMeaning
OnlineThe device and all its modules are connected and reporting
PartialThe device is connected but some functionality may not be available
OfflineThe device is not connected to the portal
UnknownThe device has never reported in (e.g., freshly enrolled)

Device detail view​

Click any device in the list to open its detail view. The detail view contains several tabs:

TabDescription
OverviewSummary cards showing outputs, network info, and hardware stats
SkillsSkills currently deployed on this device
Connected CamerasCameras linked to this device (vision-family devices only) — see Connected Cameras
SettingsDevice-level configuration including name, location, tags, and environment variables — see Device Settings
Device LogsLogs generated by the device and its skills — see Device Logs

Overview tab​

The Overview tab is divided into cards:

  • Outputs — Live values from skills currently running on the device, including health indicators.
  • Network Connections — IP address, MAC address, last boot time, and connection history.
  • Hardware — CPU usage, memory usage, disk usage, internal temperature, and audio connection status. Includes a Reboot Device button.

Connected cameras​

The Connected Cameras tab is available on vision-family devices and lists all cameras linked to the device.

To add a camera:

  1. Open the Connected Cameras tab on the device detail view.
  2. Click Add Camera.
  3. Enter the Camera Name and IP Address / URL.
  4. Click Save.

See Connected Cameras for full details.

In this section​