Closed BuddahKat closed 11 months ago
I think clipper tries to use as many threads as possible without the --processors flag, so the way to speed this up would be to have more processors available. For example, I've noticed a substantial increase when running Clipper on a 16 ppn and a 32 ppn node.
Is there a multi-threading option for the 'merge_peaks' pipeline? Just wondering if it's possible to speed it up a bit, it took 24 hours to complete and my BAM files aren't that big. I think the bottleneck is the IDR component which seems to call 'clipper' repeatedly, I know clipper is able to be run in parallel so is the anyway to pass the '--processors' flag to clipper without having to modify a bunch of files?