You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
One of the packages I have developed (ukbnmr) uses data.table and has recently run afoul of CRAN's new checks on CPU usage, which appears to have been caused in part by data.table spawning >2 threads in some of the example code.
After some discussion on the R-package-devel mailing list, I managed to address the CRAN NOTE by explicitly running setDTthreads(1) when running the examples.
The root cause of the problem seems to be a misconfiguration of the CRAN debian server, which may be missing the relevant environment variable that tells getDTthreads() to not use all available CPUs.
I wanted to draw your attention to this, as this is may potentially cause problems for other packages using data.table as a dependency.
One of the packages I have developed (ukbnmr) uses data.table and has recently run afoul of CRAN's new checks on CPU usage, which appears to have been caused in part by data.table spawning >2 threads in some of the example code.
After some discussion on the R-package-devel mailing list, I managed to address the CRAN NOTE by explicitly running setDTthreads(1) when running the examples.
The root cause of the problem seems to be a misconfiguration of the CRAN debian server, which may be missing the relevant environment variable that tells getDTthreads() to not use all available CPUs.
I wanted to draw your attention to this, as this is may potentially cause problems for other packages using data.table as a dependency.
The relevant R-package-devel thread is here: https://stat.ethz.ch/pipermail/r-package-devel/2023q3/009454.html
The text was updated successfully, but these errors were encountered: