Replies: 2 comments 1 reply
-
|
Excellent point. I've updated the blog post with a new Section 'Issue 3: detectCores() may return too many cores' covering this problem. Thank you so much. |
Beta Was this translation helpful? Give feedback.
0 replies
-
|
BTW, your comment made me think if we can automate this even further in ncores <- availableCores(constraints = "connections")I've created futureverse/parallelly#91 to track this idea. UPDATE: This was implemented in parallelly 1.33.0 (2022-12-14). |
Beta Was this translation helpful? Give feedback.
1 reply
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Uh oh!
There was an error while loading. Please reload this page.
-
Thank you for the great blog post (https://www.jottr.org/2022/12/05/avoid-detectcores/).
I'd like to add another reason to avoid
detectCores()in packages: it's becoming more common to encounter machines/clusters with more cores than R has available socket connections (~125).At least with
availableCores()a user can override the default value to limit the maximum number of cores returned.Beta Was this translation helpful? Give feedback.
All reactions