Monitoring and investigating situations
AIOps or ML-based Situations and event signature
The ML-based Situation uses an AI/ML-based event processing technique to identify event patterns from hundreds of raw events, filter out impactful events, automatically aggregate those events into Situations, and add a unique signature to each event type. The AI/ML algorithm uses the event signature to build unique Situations from the aggregated events.
Event signatures help identify the event pattern from an event sequence. However, while aggregating the events in to a Situation, the temporal relationship between the events is also considered. For example, consider an event sequence with CPU utilization events at 90%, 93%, and 97%. All events have the same pattern and signature. However, these are aggregated into the same Situation only if all these events fall within the same defined time window.
Review the following sections to learn how to monitor and investigate ML-based situations: