What is the recommended strategy for data exports to optimize CPU utilization and system performance, specifically concerning the choice between multiple smaller, targeted exports versus one large, comprehensive export?

From M-Files' perspective, is it generally more efficient for system resources (especially CPU) to run:

  • Multiple smaller, highly targeted exports: Each export focuses on a specific subset of data (e.g., a particular document collection type, a specific date range, or a limited set of properties).
  • One large, comprehensive export: A single export that gathers all necessary data for various reports or purposes (e.g., all document collections, a full year of historical data).
  • I don't know that we have metrics on this, but what is your concern here?  Is it on reducing system resources on the source (M-Files) system, or reducing resources on the target (DB) system?

    My suspicion is that the answer would be one larger export, running during quiet hours.  Smaller exports, even if you try to be careful about the export contents, are probably going to end up duplicating volumes of data.  Moreover: from recollection you will probably need to additionally manage the target databases more with smaller exports, as I think that each export needs its own database (as otherwise they'll try to use the same tables).

    So I suspect one large one is better.  If you can give me some more information on your specific concerns, though, then maybe I can try to get some better answer.