Euler Cheat Sheet

ssh <USER>@euler.ethz.ch

# Source GDC stack
source /cluster/project/gdc/shared/stack/GDCstack.sh
# Show all available tools
module avail
# Load a module
module load <name>         

# Show folder size of mapping
du -sh --si mapping
# Count files in raw_data     
find raw_data | wc -l          		
# Archive folder data1
tar cvzf data1.tar.gz data1    		
# List archive contents
tar ztvf data1.tar.gz 
# Extract all  		
tar xvf data1.tar.gz  
# Extract specific folder         		
tar xvf data1.tar.gz data1/raw 	
	
# Data transfer
scp -r data <USER>@euler.ethz.ch:/cluster/scratch/<USER>
scp -r <USER>@euler.ethz.ch:/cluster/scratch/<USER>/data ./
# Generate checksums
md5sum *fq.gz > md5sums.txt  
# Verify checksums  		
md5sum --check md5sums.txt     		

### Submit a job
sbatch < submit.tool.slurm.sh

### Overview of the submitted jobs 
jview

### Kill specific job
scancel <Job-ID>/<Array-ID>

### Kill all running jobs
scancel --user=$USER

srun --pty bash

#!/bin/bash
#SBATCH --job-name=tool	     #Name
#SBATCH --array=1-10%4       #Array length 1-10, 4 at the time
#SBATCH --ntasks=1           #Node (always 1)
#SBATCH --cpus-per-task=2    #2 CPUs
#SBATCH --mem-per-cpu=2G     #Memory per 2 CPU; 4G in total
#SBATCH --time=4:00:00       #Run time
#SBATCH --output=tool_%a.log #Logfile

source /cluster/project/gdc/shared/stack/GDCstack.sh
module load <tool> 
		
## Array variable (e.g. 1-10) and extract sample names from the index
IDX=${SLURM_ARRAY_TASK_ID}
SAMPLE=$(sed -n ${IDX}p sample.lst)

## Run the array e.g. SAMPLE 1-10 
tool -in ${SAMPLE}.in -out ${SAMPLE}.out 

jeffrun -r 
jeffrun -j <Job-ID>/<Array-ID>

# Efficency of specific job
jeff <Job-ID>/<Array-ID>
# Get an overview about the last 24 hours
jeff24
# Get the efficencies about the 7 days
jeff7