Telemetry data ingestion
Amazon IoT Events helps you evaluate telemetry data to detect events in equipment or a process. IoT Events ingests raw data from any device connected to Amazon IoT Core, processed data from Amazon IoT Analytics, and data from third party applications via IoT Events’ direct ingest APIs.
Amazon IoT Events evaluates multiple telemetry inputs to detect events and derive the state of processes, equipment, or products by applying user-defined, conditional logic. Events are patterns of data identifying more complicated circumstances than expected, such as changes in equipment when a belt is stuck or motion detectors using movement signals to activate lights and security cameras. You can schedule maintenance and send alarms or alerts prior to device failure. If you have many devices using the same detector definition, you can easily launch a new instance of the detector for every unique value of the signal. You can automatically extend your event detection and action capabilities whenever a new device comes online allowing you take advantage of stateful complex event detection at scale. Events can be detected with simple’ if-then-else’ statements applied to incoming sensor data streams.
Integration with analytics tools and other Amazon Web Services services
Amazon IoT Events can leverage output from advanced analytics services to make better decisions. With easy integrations to and from other Amazon Web Services services, you can further optimize operations. You can complete an event detector setup in Amazon IoT Events, write your event logic using simple ‘if-then-else’ statements, and select the alert or custom action to trigger when the event occurs. Such conditional statements can receive inputs from either raw data or data first processed by Amazon IoT Analytics (where you can apply machine learning models to aid in detection of more complex events). You can select a pre-built action to trigger [on the basis of a detected event], such as sending a notification through Amazon Simple Notification Service (SNS), a fully managed pub/sub messaging service, to a service technician, or you can build a custom action by setting up a function for Amazon Lambda, a compute service that lets you run code without provisioning or managing servers, to send the event data to a third-party application.
Amazon IoT Events easily scales when you are connecting many devices. Define a model once for a specific device, and the service will automatically scale and manage all devices of that model that connect to IoT Events. If an event is detected on one device or many, IoT Events can trigger the appropriate reaction or alert.
Amazon IoT Events can evaluate equipment behavior or identify equipment performance issues based on industrial data in the cloud. For an asset data property that you want to monitor, you can define an alarm rule to apply (e.g. rotations per minute is greater than a user defined value), select the severity for this alarm definition (e.g. severity values of 1, 2, 3 and 4), and configure the notifications to send when an alarm is triggered (e.g. Email and SMS). Once an alarm has been defined, operators can manage the alarm workflow by taking actions to acknowledge, snooze or disable the alarm. You can also configure additional actions to other Amazon Web Services including Amazon Lambda, Amazon Simple Queue Service (SQS), and Amazon Simple Notification Service (SNS), to be executed when an alarm triggers. This allows you to integrate alarm notifications with your own ticketing or notification system.