Geographic Information Systems Stack Exchange is a question and answer site for cartographers, geographers and GIS professionals. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

I vaguely remember being taught not to use a Feature Dataset to make things "look neat" in the database, and that the Feature Dataset should only be used when there was some sort of spatial relationship between the component feature classes.

The help file lists "organize thematically related feature classes" as a valid reason for using a feature dataset.

Is it a bad idea to use a feature dataset to group layers in the geodatabase, when the only purpose is to help navigation in ArcCatalog? Is there a performance hit from storing feature classes in a feature dataset?

share|improve this question
up vote 6 down vote accepted

Only if you plan to have users easily editing data in the feature-datasets. The biggest kick you will have is record locking; since editing a layer in a feature dataset will close out users from making other changes there.

That is probably your biggest functional killer, if that isn't a issue; then yes you can use it to try and make things pretty.

share|improve this answer
Does this mean that if you are editing, then the feature datasets are actually an advantage, since only those featureclasses within the feature dataset will be locked (without feature datasets, all featureclasses would be locked)? – Stephen Lead Jun 28 '11 at 1:49
Is this documented in ESRI help anywhere? Can you please post a link for the same. – ujjwalesri Jun 28 '11 at 4:27
The Schema locking is well documented issue; to prevent changes to a dataset while changes are happening at the same time. This is a long existing issue; for the most part it is by design to maintain data integrity...… Search ESRI site for Feature Dataset Locking – D.E.Wright Jun 28 '11 at 5:25
Thanks for the link! The reason I asked for it was because the statement The biggest kick you will have is record locking; since editing a layer in a feature dataset will close out users from making other changes there., I think is incorrect. What you might wanna say is in case of "schema edits" the dataset is "locked" to a single user, hence disabling multiple users making simultaneous changes to the schema. Record level locking would not occur! – ujjwalesri Jun 28 '11 at 6:36
Only for Personal and File Geodatabases - SDE Geodatabase can be editing with multiple users (Enterprise - Unlimited editors, Workgroup - 10 editors and Desktop - 1 editor 3 readers.)… – Mapperz Jun 28 '11 at 13:56

Grouping, in absolute terms, is not good and bad, it depends a lot on your data design and use of data.

Feature datasets are provided to allow you group spatially related FCs. The categorization, however, depends a lot on how you use the data. For example, say, you have the world data for "Streets" and "Places to Visit". This data can be grouped either by the categories identified by names above (covering all the data in just two categories), or by region i.e Asia, Europe etc.

If the purpose is to allow easy navigation in ArcCatalog, I think a good naming convention of the FCs should help there too.

Regarding the editing aspect pointed out by @D.E.Wright, I am not too sure about it. Need to check that.

share|improve this answer
Here is the link you request, is based on Schema locks to maintain data integrity when multi-users are in the system.… - this is a long known drawback and as you mention; a well planned naming convention can do just as well without the limitations. ESRI really should look to ArcSDE visualization/organization since it all really is DB objects; there is more they can do to support that. – D.E.Wright Jun 28 '11 at 5:26

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.