I believe your submission script requires an instruction to the SGE queuing system to indicate that you are indeed submitting an array job. For example, if you have 10 files (i.e., 10 tasks), you could add a line that looks like
#$ -t 1-10 SGE_TASK_ID takes on the value 1 for the first task, 2 for the second task and so on. Best regards, g -- Gowtham, PhD Director of Research Computing, IT Adj. Asst. Professor, Physics/ECE Michigan Technological University P: (906) 487-3593 F: (906) 487-2787 http://it.mtu.edu http://hpc.mtu.edu On Wed, 2 Sep 2015, VG wrote: | HI everyone, | | I am trying to make a job array script. I have 2 files in my working directory namely | abc.fastq and def.fastq | | I want to run one command on these 2 files(later on I will include more .fastq files) on | different compute nodes. So I made this array script, but it only produces output for | abc.fastq. | | Here is the script: | | #!/bin/bash | | file=$(ls -1 *.fastq | tail -n +${SGE_TASK_ID}| head -1) | filename=${file%.fastq} | awk 'NR % 2 == 0{print substr($1,7,100)};NR % 2 ==1' $file > ${filename}_BR.fastq | | I submit it as qsub -cwd -j y -t 1-2 -N testjob ./script.sh | | But only abc.fastq file gets processed?: | | What am I doing wrong here? | | Hope to hear sonn from you guys. | | Thanks | | _______________________________________________ users mailing list users@gridengine.org https://gridengine.org/mailman/listinfo/users
