A data movement service for a given business measures model is
started and stopped by starting and stopping the associated Capture and Apply
component instances. During the data movement service deployment, the start
and stop scripts have been created, so that, those scripts can be used to
start and stop the data movement services.
Your topology determines on which machines the component instances
should be running. In general, Capture components instances must run on the
machine that hosts the State database (for the State to Runtime Data Movement
Service) and on the machine hosting the Runtime database (for the Runtime
to Historical Data Movement Service). The Apply component instances should
run on the machine that hosts the Runtime database (for the State to Runtime
Data Movement Service) and on the machine hosting the Historical database
(for the Runtime to Historical Data Movement Service). In this configuration,
Apply component instances will pull data from the source database which results
in better performance than if they would reside on the machine hosting the
State database (for the State to Runtime Data Movement Service) and on the
machine hosting the Runtime database (for the Runtime to Historical Data Movement
Service).
The following information explains how to start the State to Runtime
data movement service and the Runtime to Historical data movement service.
It also describes how to stop the State to Runtime data movement service and
the Runtime to Historical data movement service.
Note: The State to Runtime
and Runtime to Historical data movement services are independent of each other.
However, it is preferable to start the State to Runtime service prior to starting
the Runtime to Historical Service. In some instances, it may be preferable
to start the Runtime to Historical data movement service after the Monitor
Server has processed entries for business measures model and
after the State to Runtime data movement service has populated the Runtime
database tables supporting this model. This can get information into the Historical
database faster than waiting for the Runtime to Historical data movement services
interval.