We have a redshift database that is writing to a PI AF table (AF updates it on periodic basis). Is it possible to detect a change in in this AF table to trigger an analysis (push vs pull?) - sorry in advance, not a database expert.
How big is this table (# records and # columns), what's the cache interval and what is the configuration of your table lookup data reference attribute ?
Ideally time series data should be written to PI Points using the PI Interface for RDBMS. You can then trigger the analyses off of PI Point updates.
It may be possible to get this to work ok with a linked table in AF, we would need a little more details and this would not be the best solution.
Is the data in your RedShift database time series data?
Retrieving data ...