Skip to main content

Hello,

As the question group suggest I am new to FME. ATM am doing some test with a .LAZ dataset (825 tiles about 60km X 60km)

1-What I want to do is read in the folder were the LAZ files are and read in a shape file (the 60x60km box).

2-clipper, to only use the LAZ tiles that are within my polygon.

3-RasterDemGenerator to create the raster in this case a DSM because there is no filter).

4-write to .TIF

The main problem here is processing time. I am new but I feel I need to do something about the processing time so I looked up what I can do about it.

1 Tiling. If I understand correctly, I can create 7 workers from a main project. So, in my case dividing this project in 2x2 or 4x4 (3x3 would already take 2 runs so logically 4x4 would be my next choice).

2 parallel processing. I tested https://community.safe.com/s/article/parallel-processing this but even on extreme I only see 2 FME Object Worker EXE in my task manager.

I have a Esri data interoperability license (2x).

A)     What I like to know If it is possible to tile my project as I described.

B)     Is it correct that I only see two FME Object Worker EXE when I use parallel processing on extreme?

C)     Is there another solution to speed up the processing time?

Using a FeatureReader with the shape as the initiator feature will allow you to only read the ones you need (your approach, with a regular reader and a Clipper, will still read all of the LAZ data before clipping)


Using a FeatureReader with the shape as the initiator feature will allow you to only read the ones you need (your approach, with a regular reader and a Clipper, will still read all of the LAZ data before clipping)

The FeatureReader roads all laz files when I set attributes on that reader and that takes a long time.


The FeatureReader roads all laz files when I set attributes on that reader and that takes a long time.

Did you set a spatial filter on the FeatureReader? (my bad, I should have mentioned that)


Reply