I am building an event-based monitoring system using GridDB. The goal is to trigger alerts when specific conditions are met in time-series data. For example:
- If a
value
exceeds a threshold for more than 5 consecutive minutes, an alert should be triggered.
Current Setup:
The data is stored in a TimeSeries container with the schema:
- timestamp (TIMESTAMP)
- sensor_id (STRING)
- value (DOUBLE)
What I Tried:
- Periodic Queries: Used Python to poll data at regular intervals and check conditions, but this approach is inefficient.
- Manual Processing: Retrieved all data and checked for threshold breaches in Python, which is not scalable.
My Question:
- Does GridDB provide a native mechanism for event-based triggering or alerting?
- If not, how can I efficiently implement a custom solution to detect and respond to threshold breaches in real time?