I've been tasked with compacting some of the personal geodatabases (MDB files) in our office every few weeks. These reside on a couple of separate file servers.

I have been using ArcCatalog to do this task, but it is very slow to launch, and switching from directory to directory also causes the title bar to display "Not Responding" for a while. All in all, the process is painfully slooooow...

However, I remember performing some compact tests on these files using Microsoft Access several months ago, and it didn't have these same problems. And the databases seemed to behave normally afterwards. However, I decided to adopt ArcCatalog because I assumed that it might be better suited to compacting these special kind of MDB files.

Was this a correct assumption? Does it matter whether I use ArcCatalog vs. Microsoft Access when I want to compact my geodatabases? Is the compact process any different between the two?

1 Answer 1


No, the assumption was not correct (ArcObjects is calling the Access function, so there should be no difference).


  1. Access-based geodatabases are "old school", with limitations in size and capability; their use has been greatly de-emphasized of late, especially in light of the deprecation of PGDB support on 64-bit platforms (no ArcGIS Server or Pro)
  2. Best practice calls for use of personal geodatabases (if they're used at all) from local disk
  3. It is exceptionally unwise to invoke the compaction option on a network share (both slow and vulnerable to catastrophic failure)

In light of this, I recommend you use a remote desktop (RDP) session to compact from the file servers locally (using whichever tool is available) and start planning transition to File Geodatabase (and at the same time work out a distribution protocol to deploy copies of a master FGDB for local use, as per best practice). If you need distributed write access, you'll need to use a storage format compatible with multi-user write access (workgroup and/or enterprise geodatabase).

Your Answer

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

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