FORUMMemory Allocation in R using Compute Canada Cluster
anandam asked 11 months ago



I am trying to run a Species Distribution Model Analysis using R in the Beluga and Cedar. One of the R objects I need to save is a large raster file. Everytime I save my R environment and try to load this raster file afterwards I get an error. It seems that important information is kept in a temporary directory and lost after the analysis is finished and I cannot load the raster object with all its metadata. One of the solutions for this problem seems to use the functions raster::readAll(rasterobject) or raster::getValues to avoid that important information is saved just temporarily. However, using these I get the error Error: cannot allocate vector of size 12.5 Gb. I am trying to increase the memory limits for R within the cluster using ulimit::memory_limit(20000), but the memory allocated does not change. My question is: HOW DO I INCREASE THE AMOUNT OF MEMORY FOR R USING R IN THE CLSUTER?

Thank you

1 Answers
zhibin Staff answered 11 months ago



How much memory did you request when you submitted your job? You can request memory for your job with “–mem”. Please refer to
 
https://docs.computecanada.ca/wiki/Running_jobs