Prediction in a distributed network

A network node, comprising:

• at least one processor;
• and
• at least one memory storing instructions that, when executed by the at least one processor, cause the network node at least to perform:
• receiving a request, from another network node, for predicted data, the request requesting activation of predicting at least one operational parameter of the network node, wherein the request comprises an indication of a trigger for triggering a prediction for generating the predicted data to be performed, the network node being caused to wait for the indicated trigger to occur prior to activating the predicting, wherein the trigger comprises:
• a time period within which the predicting should be performed,
• a plurality of time periods within which a plurality of predictions should be performed,
• at least one event for triggering the predicting, and
• an indication of a periodically repeating trigger;
• activating the predicting;
• following the activating of the predicting, generating a response to the another network node indicating the predicting has commenced, the response including an indication of a time at which the predicting started;
• receiving a prediction status request message indicating that prediction data is required every period p;
• transmitting, at every period p, a prediction status update message;
• and
• transmitting a prediction failure response in response to the predicting being unable to generate the requested predicted data, wherein the prediction failure response includes an indication of a reason for failure, the reason for failure being insufficient resource.