I am looking at using Event Pipes extensively in an application and had some questions on potential limitations of using it.
The application is driven by an AF database. An element has various attributes which are either linked to PI Tags, or to AF Lookup tables or simply Attributes which can be changed by the user. When any of these 3 parameters (PI Tag Data, AF Lookup and AF Attribute) change I need to carry out multiple calculations using all of the data linked to the AF Element.
I am expecting roughly 6000 AF Elements with multiple attributes to be monitored. Considering the engine will be monitoring on this scale and the complexity of the calculations, I am thinking of creating event pipes which are specific to each element but this means that there will be a huge number of event pipes created by the application. Is it advisable to do this?
The alternative is to collect all the PI tags into a pointlist and monitor all PI tags in one event pipe but the complexity with this is then managing which tags trigger which calculation.
Looking forward to some feedback.