A simple way to use data in value streams to build TWA models.
A let's say, there's 100GB of realtime machine data in a value stream (using InfluxDB as a persistance provider)... I want to use that data to build a model in TWA. You find these data quantities when you have sampling in the seconds and store years of data. The benefit of having years of data is being able to see seasonal effects on machines e.g. machine in Scandinavia operating between -30°c and +30°c ambient temp. It also produces crazily accurate models