Complex Event Processing
AggreGate Event Correlator is an integrated powerful tool for Complex Event Processing. It is based on open-source Siddhi library that is continuously improved.
While bindings are widely used in AggreGate as a way to react single events, you can use Event Correlator to manipulate a group of incoming events: sequence, time window, several streams and so on. Event chains are processed on the fly keeping them in memory. Complex Event Processing engine doesn’t make unnecessary operations of writing and reading from persistent storage.
Numerous ways of Event Correlator implementation are available. Here are some of them:
Predefined patterns detection | High-frequency signal downsampling |
Trends tracing | Full and incremental aggregation over time, length, and session windows |
Correlation of several event streams | Detection of anomalies and missing events |
Deduplication of identical events | Real-time predictions with online and pre-trained Machine Learning models with PMML support |
Throttling and debouncing | Connection of external event flows |
Event Correlator supports not only internal AggreGate Unified Data Model events. Complex Event Processing can be performed on mixed internal and external event sources and sinks including:
- Kafka
- JMS/MQ
- NATS
- CDC
- Files
- and more
Default package of Event Correlator contains most necessary plugins and Unified Data Model connectors. Other Siddhi plugins can be easily added to specified extensions folder.
Due to unified architecture all power of Complex Event Processing is available not only in Cloud or On-premise installations. Event Correlator is a light-weight module, included in AggreGate Edge. Edge device can execute complex business logic itself. You can send only necessary data to central managing node. So, using Complex Event Procesing on Edge tier gives a lot of advantages by increasing autonomy of data processing and reducing traffic, latency, storage requirements.
Power of Event Correlator and Distributed Architecture capabilities reveal new opportunities in all industries when we deal with many data streams and need complex analytics and decision making on the fly.