Closed paul-shannon closed 3 years ago
I had a problem with how I was implementing the parallel code, It should be fixed here (on github) now, and should be propagating to Bioconductor devel soon.
Thanks, Simon!
I’m trying it out now.
On Sep 10, 2021, at 12:42 PM, Simon Coetzee @.***> wrote:
I had a problem with how I was implementing the parallel code, It should be fixed here (on github) now, and should be propagating to Bioconductor devel soon.
— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub, or unsubscribe. Triage notifications on the go with GitHub Mobile for iOS or Android.
Works great! Fast!
Thanks, Simon.
On Sep 10, 2021, at 12:56 PM, Paul Shannon @.***> wrote:
Thanks, Simon!
I’m trying it out now.
- Paul
On Sep 10, 2021, at 12:42 PM, Simon Coetzee @.***> wrote:
I had a problem with how I was implementing the parallel code, It should be fixed here (on github) now, and should be propagating to Bioconductor devel soon.
— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub, or unsubscribe. Triage notifications on the go with GitHub Mobile for iOS or Android.
@scoetzee @Simon-Coetzee First, thanks for this package. It has been very helpful.
In scaling up our use to explore eQTLs from Alzheimer's loci, I cannot seem to make good use of the BPPARAM argument.
With
MulticoreParam(workers=40)
on a linux box with 84 cores, watchingtop
I see multiple R processes appear very briefly, and then disappear, leaving only one R process, running for an additional 10 minutes. The total execution time seems to be linear.I am using 40 rsids at a time, against 80 motifs from MotifDb.
Any suggestions?