You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Copy file name to clipboardExpand all lines: articles/iot-operations/discover-manage-assets/overview-akri.md
+50-32Lines changed: 50 additions & 32 deletions
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -1,6 +1,6 @@
1
1
---
2
-
title: Discover assets (preview)
3
-
description: Understand how the Akri services enable you to discover devices and assets at the edge, and expose them as resources on your cluster.
2
+
title: Learn about Akri (preview)
3
+
description: Understand how the Akri services enable you to dynamically configure and deploy Akri connectors to connect a broad variety of assets and devices to the Azure IoT Operations cluster, ingest telemetry from them, and use command and control.
4
4
author: dominicbetts
5
5
ms.author: dobett
6
6
ms.subservice: azure-akri
@@ -12,57 +12,71 @@ ms.date: 07/08/2025
12
12
# CustomerIntent: As an industrial edge IT or operations user, I want to to understand how the Akri services enable me to discover devices and assets at the edge, and expose them as resources on a Kubernetes cluster.
13
13
---
14
14
15
-
# What is asset and device discovery (preview)?
15
+
# What are Akri services (preview)?
16
16
17
-
Azure IoT Operations discovers devices and assets by using the included Akri services (preview). The Akri services enable protocol connections and configurations by using Azure Device Registry. The Akri services simplify the process of projecting leaf devices such as OPC UA servers, cameras, IoT sensors, and other assets into the Azure Device Registry. The Akri services use the devices' own protocols to project leaf devices as Azure Resource Manager resources in the Azure Device Registry. For administrators who attach or remove devices from a cluster, this capability reduces the amount of coordination and manual configuration required.
17
+
The Azure IoT Operations southbound connectors use Akri services to:
18
18
19
-
The Akri services are an extensible framework for all device protocols. You can use them with out-of-the-box and partner-built connectors, or you can add custom discovery and provisioning capabilities by adding protocol handlers, connectors, and behaviors. Adding custom logic is made easy with the [Azure IoT Operations SDK](https://github.com/azure/iot-operations-sdks) (preview).
19
+
- Discover physical devices and devices connected to your cluster.
20
+
- Enable connectivity to the physical assets and devices by using protocols such as OPC UA, ONVIF, and REST/HTTP.
21
+
- Configure namespace assets and devices as custom resources in your Kubernetes cluster.
22
+
- Integrate with Azure Device Registry to project device and namespace assets to the cloud as Azure Resource Manager resources, reducing the amount of manual configuration required.
23
+
24
+
The Akri services provide an extensible framework for all device connectivity protocols. The following types of southbound connector all use Akri services:
25
+
26
+
- Built-in connectors such as the **connector for OPC UA** and **connector for REST/HTTP (preview)**.
27
+
- Partner-provided connectors.
28
+
- Custom connectors.
29
+
30
+
The [Azure IoT Operations SDK (preview)](https://github.com/azure/iot-operations-sdks) includes examples to help you get started building custom connectors. The SDK samples show you how to:
31
+
32
+
- Use the Akri services to deploy connectors and manage their lifecycle.
33
+
- Interface with other Azure IoT Operations components such as the MQTT broker.
20
34
21
35
The Akri services are a Microsoft-managed commercial version of [Akri](https://docs.akri.sh/), an open-source Cloud Native Computing Foundation (CNCF) project.
22
36
23
37
## Leaf device integration challenges
24
38
25
-
It's common to run Kubernetes directly on infrastructure. But to integrate non-Kubernetes IoT leaf devices into a Kubernetes cluster requires a unique solution.
39
+
In Azure IoT Operations, your Kubernetes cluster runs on your edge infrastructure, which introduces challenges when you want to integrate non-Kubernetes IoT leaf devices. For example:
26
40
27
-
IoT leaf devices present the following challenges, They:
28
-
29
-
- Contain hardware that's too small, too old, or too locked-down to run Kubernetes.
30
-
- Use various protocols and different topologies.
31
-
- Have intermittent downtime and availability.
32
-
- Require different methods of authentication and secret storage.
41
+
- They use hardware that's too small, too old, or too locked-down to run Kubernetes.
42
+
- They use various protocols and different topologies.
43
+
- They have intermittent downtime and availability.
44
+
- They require different methods of authentication and secret storage.
33
45
34
46
## Core capabilities
35
47
36
-
To address the challenge of integrating non-Kubernetes IoT leaf devices, the Akri services have several core capabilities:
48
+
To address the challenges of integrating non-Kubernetes IoT leaf devices, the Akri services have several core capabilities:
37
49
38
-
### Device discovery
50
+
### Connector deployment and lifecycle management
39
51
40
-
Akri services deployments can include fixed-network discovery handlers. Discovery handlers enable assets from known network endpoints to find leaf devices as they appear on device interfaces or local subnets. Examples of network endpoints include OPC UA servers at a fixed IP address, and network scanning discovery handlers.
52
+
Akri services include the _Akri operator_. The operator lets you deploy connectors dynamically when certain types of devices are found on the cluster and the corresponding namespace assets are allocated to the connector. The Akri operator provides automatic access to Azure IoT Operations resources such as device endpoints and namespace assets.
41
53
42
-
### Dynamic provisioning
54
+
### Asset discovery
43
55
44
-
Another capability of the Akri services is dynamic device provisioning.
56
+
Akri services include the _Azure Device Registry service component_, which works with the connectors that have discovery capabilities. These connectors enable the metadata and definitions such as datasets and events to be onboarded through known device endpoints. The Azure Device Registry service component creates the discovered namespace assets as custom resources that an OT user can view in the operations experience web UI. The OT user can then onboard the discovered assets as namespace assets that are automatically added to the Azure Device Registry.
45
57
46
-
With the Akri services, you can dynamically provision devices such as:
58
+
### Physical device discovery
47
59
48
-
- USB cameras to use in your cluster.
49
-
- IP cameras that you don't want to look up IP addresses for.
50
-
- OPC UA server simulations running on your host machine that you use to test Kubernetes workloads.
60
+
Akri services deployments can include fixed-network discovery handlers. Discovery handlers enable assets from known network endpoints to find leaf devices as they appear on device interfaces or local subnets. Examples of network endpoints include OPC UA servers at a fixed IP address, and network scanning discovery handlers. This capability isn't supported in the current release.
51
61
52
62
### Compatibility with Kubernetes
53
63
54
-
The Akri services use standard Kubernetes primitives that let you apply your existing expertise and knowledge. Small devices connected to an Akri-configured cluster can appear as Kubernetes resources, just like memory or CPUs. The Akri services controller enables the cluster operator to start brokers, jobs, or other workloads for individual connected devices or groups of devices. These device configurations and properties remain in the cluster so that if there's node failure, other nodes can pick up any lost work.
64
+
The Akri services use standard Kubernetes primitives that let you apply your existing expertise and knowledge:
65
+
66
+
- Small devices connected to an Akri-configured cluster can appear as Kubernetes custom resources, just like memory or CPUs. These device and asset configurations and properties remain in the cluster so that if a node fails, other nodes can pick up any lost work.
55
67
56
-
## Connectors supported
68
+
- Security and authentication use standard Kubernetes secrets and TLS practices, which makes it easy to secure your device connections regardless of the connectivity protocol.
69
+
70
+
## Supported connectors
57
71
58
72
The following table shows the connectors currently available in Azure IoT Operations and their asset and device discovery capabilities:
| Dynamic discovery of devices at the edge (supported protocols: OPC UA, ONVIF, udev) | Yes |
81
95
| Schedule devices with minimal latency using Akri's information on node affinity on the cluster | Yes |
82
-
| View Akri metrics and logs locally through Prometheus and Grafana | Yes |
83
96
| Secrets and credentials management | Yes |
84
97
| M:N device to broker ratio through configuration-level resource support | Yes |
85
-
| Observability on Akri deployments through Prometheus and Grafana dashboards | Yes |
86
98
87
99
| Akri services features | Supported |
88
100
|--------------------------|:---------:|
89
-
| Installation through the Akri services Arc cluster extension | Yes |
101
+
| Installation through the Azure IoT Operations Arc extension | Yes |
90
102
| Onboard devices as custom resources to an edge cluster | Yes |
91
103
| View the Akri services metrics and logs through Azure Monitor | Yes |
92
104
| The Akri services discover and create assets that can be ingested into the Azure Device Registry | Yes |
93
105
| Akri services configuration by using the operations experience web UI | Yes |
94
-
| Deployment and management features for integrating non-Microsoft or custom protocol connectors and discovery handlers | No |
106
+
| Deployment and management features for integrating non-Microsoft or custom protocol connectors | Yes |
107
+
| Deployment and management features for integrating non-Microsoft or custom protocol discovery handlers | No |
95
108
96
109
## Related content
97
110
98
-
To learn more about the Akri services, see [Discover OPC UA data sources using the Akri services](howto-autodetect-opc-ua-assets-use-akri.md)
111
+
To learn more about OPC UA automatic asset discovery with Akri services, see [Discover OPC UA data sources using the Akri services](howto-autodetect-opc-ua-assets-use-akri.md)
112
+
113
+
To learn more about using Akri with ONVIF, Media, or REST/HTTP, see:
114
+
-[Understand the connector for media](./overview-media-connector.md)
115
+
-[Understand the connector for ONVIF](./overview-onvif-connector.md)
116
+
-[Understand the connector for REST/HTTP](overview-http-connector.md)
99
117
100
118
To learn more about the open-source CNCF Akri, see the following resources:
Copy file name to clipboardExpand all lines: articles/iot-operations/discover-manage-assets/overview-manage-assets.md
+6-8Lines changed: 6 additions & 8 deletions
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -4,7 +4,7 @@ description: Understand concepts and options needed to manage the devices and na
4
4
author: dominicbetts
5
5
ms.author: dobett
6
6
ms.topic: overview
7
-
ms.date: 07/17/2025
7
+
ms.date: 07/28/2025
8
8
ai-usage: ai-assisted
9
9
10
10
# CustomerIntent: As an industrial edge IT or operations user, I want to understand the key components in the Azure IoT Operations for managing devices and assets, so that I can effectively manage the devices and assets in my solution.
@@ -34,7 +34,7 @@ In the previous diagram:
34
34
- The assets, such **Asset-01** that could be an oven, are *physical assets* that connect indirectly through one of the OPC UA servers.
35
35
- The OPC UA servers are *physical devices* that connect to Azure IoT Operations through the connector for OPC UA.
36
36
37
-
A physical device can connect using a variety of protocols. It could connect through a connector such as the media connector. If it uses the MQTT protocol, it can connect directly to the MQTT broker bypassing the connectors.
37
+
A physical device can connect using various protocols. It could connect through a connector such as the media connector. If it uses the MQTT protocol, it can connect directly to the MQTT broker bypassing the connectors.
38
38
39
39
The following diagram shows how physical devices connect to Azure IoT Operations:
40
40
@@ -89,7 +89,7 @@ Azure IoT Operations includes several services that help you manage devices and
89
89
- The **connector for ONVIF (preview)** is a service that discovers and registers ONVIF assets such as cameras. The connector enables you to manage and control ONVIF assets such as cameras connected to your cluster.
90
90
- The **connector for REST/HTTP (preview)** is a service that lets you connect to REST/HTTP endpoints and publish data to the MQTT broker.
91
91
-**Custom connectors** are services that you can create to connect to other data sources and publish data to the MQTT broker. Use the Azure IoT Operations SDKS to create custom connectors that meet your specific requirements.
92
-
-**Akri services**are a set of services that enable the automatic discovery of physical assets and devices in your environment. Akri services can help you create and configure assets and devices in operations experience. To learn more, see [What is asset discovery (preview)?](overview-akri.md).
92
+
-**Akri services (preview)** enable the automatic discovery of physical devices and assets and help OT users configure devices and namespace assets in the operations experience web UI. The connectors described previously all use the framework provided by the Akri services to implement their core capabilities.
93
93
94
94
## Store assets as Azure resources in a centralized registry
95
95
@@ -163,10 +163,8 @@ Spec:
163
163
Events: <none>
164
164
```
165
165
166
-
### Automatic asset discovery
166
+
### Protocol connectivity and asset discovery
167
167
168
-
Akri services let you deploy and configure connectivity protocols, such as OPC UA and ONVIF, at the edge. Akri services use the asset and device resources in Azure Device Registry to model the different device and protocol connections in your environment.
168
+
Akri services let you deploy and configure connectivity protocolsat the edge. Akri services use the asset and device resources in Azure Device Registry to model the different device and protocol connections in your environment. It enables you to easily onboard and provision assets with open standards. It provides an extensible framework for all device protocols and a single pane of glass view for all assets.
169
169
170
-
Akri services simplify the process of creating assets by automatically onboarding assets with pre-configured datasets and device endpoints generated by the connectors to represent capabilities and devices on the network.
171
-
172
-
To learn more, see [What is asset and device discovery (preview)?](overview-akri.md).
170
+
Akri services simplify the process of creating assets by automatically onboarding assets with pre-configured datasets and endpoints generated by the connectors to represent capabilities and devices on the network.
0 commit comments