The device diagnostics capability of IoT Platform provides intelligent services for scenarios in which you need to perform O&M tasks on the running data of devices. The capability helps you analyze and identify device failures, detect potential risks on devices, and obtain the details of business exceptions at the earliest opportunity.
Architecture
Benefits
Various data diagnostic metrics
Provides various data diagnostic metrics that can help you check devices and identify issues on the devices in a comprehensive manner. The data diagnostic metrics include Network Connection Success Rate, Cause, and Cloud Connections.
Automated intelligent baselines
Provides automated alert thresholds for metrics to identify abnormal devices based on the running data of the devices. The capability automatically generates intelligent baselines by using machine learning algorithms. This helps you identify and manage abnormal devices in an intelligent manner.
Exception detection alerts
Supports real-time monitoring on device metrics. If the value of a device metric reaches a threshold value, an alert is triggered. This helps you identify device exceptions at the earliest opportunity.
Features
Diagnostic metric
Displays the data diagnostic metrics of the current instance in a visualized manner. The data diagnostic metrics include Network Connection Success Rate, Message Arrival Rate, and Time Consumed for Device-to-cloud Connection. This helps you view and analyze the status of devices.
Intelligent baseline
Automatically generates metric-based intelligent baselines based on machine learning algorithms and checks abnormal metrics by using the baselines. This helps prevent static threshold-triggered alerts and false positives.
Offline analysis
Collects the characteristics of disconnected devices based on the time range within which the disconnection events occurred and the timeout period for device reconnection, including the number of device connection events and the distribution of device disconnection events by region. This helps you obtain the details of disconnected devices and analyze the causes.
Diagnostic analysis
Provides metric-based diagnostic services for specific devices, including Unexpected Device Disconnection, Upstream Messaging Exceptions, Abnormal Downstream Messaging, and Rule-based Message Forwarding Exceptions. This feature also provides corresponding suggestions for troubleshooting based on the diagnostic result.
Limits
The device diagnostics capability is available only in the China (Shanghai), China (Beijing), and China (Shenzhen) regions.
Billing
The device diagnostics capability includes the diagnostic metric, intelligent baseline, offline analysis, and diagnostic analysis features. If you enable one of the features, a specific number of data processing units (CUs) are consumed. For more information about consumption rules, see Enterprise Edition instances (subscription).
Process
Enable or disable device diagnostics features: Before you can view diagnostic information about an Enterprise Edition instance that you purchased, you must enable the required device diagnostics features for the instance.
For more information about how to use the features of the device diagnostics capability, see the following topics:
View the CU usage: describes how to view the status of the diagnostic metric, intelligent baseline, offline analysis, and diagnostic analysis features, and the CU usage of the current instance.