Delay when querying cached features from transformer

  • 20 January 2024
  • 1 reply

When enabling feature caching, there is a 'considerable' lag when trying to interrogate the cached features from an output port: usually have to wait 10 -15 seconds to expand the table attribute fields or sort the attribute fields, for instance. When clicking elsewhere to do another activity and returning to the cached features output, the same lag time is experienced. I though the idea was that cached features would be stored virtually, so they wouldn't have to be re-loaded each time? The lag is even present when only interrogating a handful of features. Are there settings that can optimise this to reduce the lag/remove? I do use a VPN to access the workspaces from OneDrive (there workspaces are not stored locally).


Version: 2022.2.3.0 (20230131 - Build 22789 - WIN64) 

1 reply

Userlevel 3
Badge +24

The VPN could be a factor here, the fact that the workspaces are on OneDrive probably not so much.


FME stores the data for the feature caches in memory or in temp files that are located in the location specified by the FME_TEMP environment variable or, if that is not set, your OS's default temp file location. If you have the option of putting that on a faster hard drive (preferably an SSD) it might make a difference, but my bet would be on the VPN.