[gdal-dev] GDAL python bindings memory usage

[Even Rouault]

On jeudi 26 décembre 2019 14:30:35 CET Evert Etienne (SITEMARK) wrote:
> Using a bigger file (8GB) and a machine with 64GB Ram we can see the
> increase being higher. For both gdal.Warp and gdal.Translate

GDAL may use up to GDAL_CACHEMAX memory, plus some other smaller caches. I'm 
not sure how memory_profile monitors memory consumption, but it is possible 
that the apparent virtual memory is not returned to the OS. But if you run 
those commands several times in the same procss, you should hopefully observe 
some ceiling in memory consumption.

Even

-- 
Spatialys - Geospatial professional services
http://www.spatialys.com