It appears that vsearch is ignoring the # of processors set in mothur.
Using vsearch agc method for clustering, I am seeing the following (actual utilization is lower than 32, thankfully, likely due to disk IO):
Current files saved by mothur:
fasta=stability.trim.contigs.good.good.unique.good.good.filter.precluster.pick.fasta
name=stability.trim.contigs.good.good.unique.good.good.filter.precluster.pick.names
processors=8
Current default directory saved by mothur: /usr/analysis/tools/mothur/
Current working directory: /gluster/scholzmb/AllSamples/data/temp/
Output File Names:
current_files.summary
mothur > cluster(fasta=current, name=current, method=agc) Using stability.trim.contigs.good.good.unique.good.good.filter.precluster.pick.fasta as input file for the fasta parameter. Using stability.trim.contigs.good.good.unique.good.good.filter.precluster.pick.names as input file for the name parameter. You did not set a cutoff, using 0.03. vsearch v2.0.2_linux_x86_64, 378.5GB RAM, 32 cores https://github.com/torognes/vsearch