Pattern

Pattern Nodes in Cortex serve as state machine implementations for Complex Event Processing (CEP), enabling the detection of predefined event patterns across one or more event streams over time. These nodes are adept at matching patterns, counting event occurrences, and managing logical event ordering with conditions like "and," "or," and "not."

Purpose: Pattern Nodes are designed to identify complex event patterns in real-time, without requiring immediate succession of matching events. They efficiently correlate events that are temporally separated, ensuring comprehensive pattern detection.

1. States Creation: Users can define multiple States within a Pattern Node, each acting as a component of the state machine. In every State, there's the option to include up to two occurrences, non-occurrences, and counting conditions.

2. Conditions and Time Limits: For each State, you can specify conditions including occurrence, non-occurrence, and counting. Time limits can be set for non-occurrences within each State, as well as an overall time limit for the Pattern Node, enhancing control over event processing timing.

3. State Progression: States capture and process events from Stream Nodes. The progression from one State to the next is based on the evaluation of conditions within the current State. If the conditions return True, the State machine moves to the next State. Successful progression through all States, with each returning True, leads to the Pattern Node relaying output attributes from the last State to the subsequent Node.

Pattern Nodes play a crucial role in Cortex's CEP capabilities, allowing for intricate event pattern detection and processing, thereby enabling sophisticated real-time data analysis and decision-making processes.