🧬 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          

srun --pty bash

#!/bin/bash
#SBATCH --job-name=tool                 #name
#SBATCH --array=1-3%3                   #array length 1-3
#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 <name> 
        
## Array variable (e.g. 1-3) and extract sample names based on index from
IDX=${SLURM_ARRAY_TASK_ID}
SAMPLE=$(sed -n ${IDX}p sample.lst)

## Run the array e.g. SAMPLE 1-3 
tool -in ${SAMPLE}.in -out ${SAMPLE}.out 

👩‍💻
### 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

jeffrun

# get an overview about the last 24 hours.
jeff24

### On Euler
get_inefficient_jobs