Yeah, the reason I went with disabling virtualization was precisely because I knew the datasource is mostly going to be a fixed size. Enabling the virtualization made a big impact on the initial load time, but unfortunately it didn't help as much with the adding/removing of columns.
I have actually been playing around with column visibility already, and it did cut down quite a bit of time, hence why I'm considering declaring the table statically with a maximum number of columns then hiding them. The batch update seems like it could actually work well in our case since the adding/removing of columns normally happen all at once (maybe 30~50 columns at a time or more).
Thanks!
Alex
I have actually been playing around with column visibility already, and it did cut down quite a bit of time, hence why I'm considering declaring the table statically with a maximum number of columns then hiding them. The batch update seems like it could actually work well in our case since the adding/removing of columns normally happen all at once (maybe 30~50 columns at a time or more).
Thanks!
Alex