Category: R

How to submit simple jobs in PICB cluster

ls job*.sh | xargs -n1 qsub -q linux.q

How to run R jobs in parallel in PICB cluster with SGE system

qsub.sh

#!/bin/sh
mkdir ./sge.out
chmod u+x *

for i in `seq 1 100`; do
  qsub -q linux.q test.sge $i;
done

test.sge

#!/bin/sh

# Change to the current working directory upon starting of the job
#$ -cwd

# run job on the node with at least a certain size of free memory
#$ -l mem_free=8G

# numbers of slots for this job, minimum 2, maximum 4
#$ -pe orte-test 2-4

# do not flood myself with e-mail
#$ -m n

# join the error and standard output streams
#$ -j yes

# Defines or redefines the path used for the standard error  stream and the stan
dard output stream of the job. 
#$ -e ./sge.out/
#$ -o ./sge.out/

# name of the job
#$ -N test_sge_system_in_picb

/picb/home5/augix/bin/R/bin/Rscript ./test.R $1

test.R

args <- commandArgs(TRUE)
flag<-args[1]
print(flag)
write.table(flag,file=paste(flag,".txt",sep=""))
q(save="no")

See Also

SGE in eva cluster