Question

Clipping from large FGDB

  • 13 August 2013
  • 2 replies
  • 11 views

Morning All

 

 

i currently need to extract OS data by polygon extents from a fgdb. Setting the polygon is not a problem, and clipping the data to this polygon works. The problem is that there are 6 readers, each reading in different OS features e.g. Area, Line, Boundary etc. The line readers reads in 64000000 features which crashes my FME and also takes an age.

 

 

I know what i'm doing works as i can extract from the smaller feature readers, but is there a different transformer i should be using rather than clipper? or before the clipper to avoid reading all the data into the workspace? The reading of all the data seems to be the problem as it's reading in over 100,000,000 OS features.

 

 

any help of guidance is greatly appreciated.

 

 

Chris

2 replies

Userlevel 4
Badge +13
Hi Chris,

 

 

There are several possibilities to limit the amount of features:

 

  1. Use the extent polygon as the initiator in the feature reader transformer(s) and set the spatial interaction to intersect.
  2. An alternative would be to extract the boudingbox coords from the the extent polygon and use them in the gdb reader min/max parameters.
  3. And finally you can use the polygon as a search feature in the gdb reader (some pre-processing necessary)
Hope this helps.
Userlevel 4
Badge +13
forgot to mention that there is a webinar specifically for improving preformance

Reply