Skip to main content

Hi all. First post for me here.

I have a workbench where I want to overlay a point dataset on a single band geotiff elevation raster dataset. This workbench will be published to FME server where ppl will uploaded their point dataset and terrain Info will be calculated.

The GeoTiff dataset is large >10 gb. To speed up the process I have tiled the dataset into smaller tiles and created a shape file index. I only want to load the tiles which overlap the point dataset.

I find out which tiles overlap the point dataset and then, using a feature reader transformer, I’m selecting the tile to load. My question is, what would be the best way to load multiple tiles if the point dataset overlaps more than one tile.

Is it best to use multiple feature readers with each tile being passed to one feature reader. But I’ll never know how many feature readers I need.

Any thoughts. TIA

TK

One FeatureReader should be enough, even if you input multiple points, it will read all the GeoTiff's necessary. You may want to aggregate the points though, to avoid a GeoTiff being read more than once if there's more than one point overlapping.


One FeatureReader should be enough, even if you input multiple points, it will read all the GeoTiff's necessary. You may want to aggregate the points though, to avoid a GeoTiff being read more than once if there's more than one point overlapping.

Oh thank you. I haven’t used the spatial filter option with the feature reader before. If I select all of the tiles and then use the overlap spatial filter it works well. Thanks.


Reply