The Pricefx server has a so called heartbeat service that at given intervals performs common tasks. The default interval is 1 minute; it is configured in pricefxin pricefx-config.xml xml (backgroundWorkers/pollingPeriod). The calculation flow processor is hooked to this heartbeat service, so every minute the processor checks the definitions of calculation flows and if the time has come, it loads the definition of a given calculation flow and processes it.

...

To enable it, use the ServerRole: "Calculation Flow Failover Processor" for the node(s) which should pick up processing if the master node (i.e., a node with "Calculation Flow Processor") fails.

CF Calculation Running on Cloud Provider Backend Node

...

• For individual CFs by enabling the option Disable on service role in the CF configuration. This is a preferred option if you want to keep your costs down because you are running only the required jobs inside pods. The next two options are more costly because running a CF as a pod is more expensive than running it on the backend node.
  Note: If the disableCFOnServiceRole property is enabled on the cluster or partition, all CFs will run on a dedicated pod regardless of individual CF settings.

• For the current partition by adding the Advanced Configuration Option disableCFOnServiceRole and setting its value to true.

• Cluster-wide by setting the property disableCFOnServiceRole to true in the pricefx-config.xml file.
  Note that only the support team can change the server configuration.

You can find out in the UI if a Calculation Flow job is running on a service role or a dedicated pod. In Jobs & Tasks or Calculation Tracking, make the Lightweight column visible and check the value for your Calculation Flow. If the value is true, it runs on the backend node. If the value is false, it runs on its dedicated pod.

...