Hello,
We have several views in a separate SQL Server enterprise database that we are using to map points in our enterprise geodatabase(s) for different types of data. We are copying the data from the views into our EGDB as a static feature class with points created and selected columns (due to ESRI not playing nice and efficiently with using data across networks and servers).
Our question is: should we bring the entire view (table) over and use definition queries in the mxd (project) to publish map layers or should we use sql to create individual tables to be brought in to the mxd (project) before publishing the map layers? Is there a cost or benefit (space, time, messiness, etc) to either way? This process will be run automatically on at least a nightly basis, possibly twice a day when decided.
Thanks,
Lorinda
Hello Lorinda,
As best practice, I would discourage the publishing of views. I haven't had much luck doing so in the past.
I have a similar workflow that derives from views, but I use scripts to migrate the data into a feature class and then I publish the feature class as a feature service. Make sense?
That is the plan, use the views in the original data to create feature classes in the SDE. Question is: should we do one feature class or break it into multiples? Performance hits for migration or display are considerations.
Thanks
Lorinda
I guess the answer to one or multiple also depends on the usage of the feature service. I would test both.