Hi,
I'm experiencing the same script to run faster on rstudio 1.2.5033 than in 1.4.1106. It seems related to the former using all the threads in the CPU while the latter only one. I'm working in a Linux Mint 20 environment of two highly similar, albeit different, machines with the same 5.4 kernel. I am not an expert and I do not understand what is going on.
Can anyone please point me to the right documentation to better understand this different behavior?
RStudio doesn't implement any parallelism, most R functions are single threaded out of the box and it is up to you to implement parallel processing. I would suspect your R setup is different in both machines (other than RStudio).
Maybe if you profile your code in both machines, that would shed light on the problem. Also, check your sessionInfo() and look for differences.
Hi again,
thanks a lot for your help, you were totally right it was the R setup. I forgot that my Linux distribution install by default generic BLAS/LAPACK libs and installing openBLAS fixed the issue. FYI R script uses a lot matrix computations so it makes a huge difference.