Time Series is a value which is written in a store, received from the specified data source (tag or expression) and identified by the address (ID). Time series are organized in groups and are presented as a hierarchical structure.
Interaction between time series and a store is performed on the model of publisher-subscriber, where publisher is a time series and subscriber is a store. It means that time series notifies the associated store about the fact that it has changed. In its turn the store accumulates the received values and writes them in the target database. Notification about changing of the time series occurs in the following cases:
- Data source (tag or expression) associated with the time series has changed.
- Data source (tag or expression) associated with the time series is not changing during a certain period of time. This mode is turned on optionally.
Please take note that time series supports only those data sources which values can be converted into double (64-bit floating point number, IEEE-754), e.g.: UInt64, Int32, Boolean, Byte, etc. Non scalar data types are not supported