I am currently working with large amounts of forest data where surveyors and ecologists measure and assess the quality of the vegetation in the field and send the result as CAD files on a daily basis to be converted, edited and stored in a large enterprise geodatabase. It's a rather large project with many people involved.

Even though I've created basemap layers from layers that do not need to be snapped against, it's slow to work. I know that in some CAD software (I only know the basics of CAD), you can load just a small extent of data from a database into the view and work only with that extent without having to render the entire database, but is it possible to do the same thing in ArcMap?

I've tested a lot of workarounds, clipped data to a file-based geodatabase, for example, and then copied the objects to the enterprise database after editing, but it gives a rather clumsy workflow.

If I define the spatial extent in the data frame, is it still working on the entire database? I'm not experiencing improved performance when doing so.

The optimum would have been to open an MXD with a WMS background map, draw a rectangle and load data from the database within the extent from the rectangle but I assume that ArcMap isn't built to work that way?

So, what is the easiest way to work with a small spatial extent of a large dataset?

  • How many featureclasses do you have in your geodatabase? What happens if you turn off all the geodb layers, then right click on the CAD layer, choose zoom to layer, then turn on the geodb layers? – Kirk Kuykendall Nov 27 '18 at 18:37
  • The CAD-file is converted to file gbd, edited and then copied to the enterprise gdb. There are usually som editing needed with snapping, topology checks and so on so it would be preffered to load all features at once. – user21070 Nov 27 '18 at 19:08
  • 2
    since you're using an enterprise db, you could create spatial indices for the cad data extents. it might be faster than clipping data and should improve the load times. – jbalk Nov 27 '18 at 19:10
  • 1
    I second the comment by @jbalk : proper (spatial) indexing should allow you to work comfortably with the entire dataset, even if it contains millions of records. Note that after you import lots of data, it may sometimes help to rebuild the spatial indexing. And finally, you may want to seta maximum scale, so that features will only be drawn when zoomed in. – Berend Nov 27 '18 at 20:05

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.