Creating batch or streaming data metrics

You can create a metric that runs a function on data and generates a more complex value or triggers an alert. For example, you can create a metric that determines when current run hours drop below a defined threshold and that triggers a dashboard alert for the related device. These metrics are also known as calculated metrics or function-based metrics.

About this task

You can create the following two types of function-based metrics:

Batch data metric
A metric that uses data that is loaded in batches from the database.
Streaming data metric
A metric that uses data that is streamed from the IoT tool. Streaming data metrics can load data and send alerts within seconds, but you must refresh the page or chart to see the latest data. Streaming data metrics were introduced in Maximo® Monitor 8.9. Streaming data metrics do not support summary dashboards.

These metrics can exist individually or as part of a pipeline. For example, you can create a batch data metric that uses a function to determine the standard deviation of temperature data. That batch data metric has temperature data as the input and standard deviation as a function. You can specify that the output of that batch data metric is the input to another batch data metric.

Maximo Monitor includes many built-in functions, including anomaly detection functions. For more information, see Built-in functions.

Procedure

  1. Select a device type that is not based on a sample device type. Sample device types do not support batch or streaming data metrics.
  2. On the Data tab, click Create metric for batch data metric or, if available, streaming data metric.
  3. Select a function to use to calculate the metric and then click Select.
  4. Set the scope for the function and then click Next.
  5. Specify the input for the metric and then click Next.
  6. Specify the output for the metric.
  7. Click Create.