benchmark.yml

name: Run Simple R Script on HPC via Slurm

on:
  push:
    branches:
      - feature/ci-cd-pipeline

jobs:
  Benchmarking-pipeline:
    runs-on: ubuntu-latest
    env:
      FDR_THRESHOLD: 0.2

    steps:
    - name: Checkout Repository
      uses: actions/checkout@v3

    - name: Set Up SSH Access
      run: |
        mkdir -p ~/.ssh
        touch ~/.ssh/id_rsa
        chmod 600 ~/.ssh/id_rsa
        echo "${{ secrets.SSH_PRIVATE_KEY }}" > ~/.ssh/id_rsa
        ssh-keyscan -H login-00.discovery.neu.edu >> ~/.ssh/known_hosts || exit 1

    - name: Transfer Files to HPC
      run: |
        # scp benchmark/benchmark.R benchmark/config.slurm raina.ans@login-00.discovery.neu.edu:/work/VitekLab/Projects/Benchmarking || exit 1
        scp benchmark/benchmark.R benchmark/config.slurm raina.ans@login-00.discovery.neu.edu:/home/raina.ans/R || exit 1

    - name: Submit Slurm Job and Capture Job ID
      id: submit_job
      run: |
        ssh raina.ans@login-00.discovery.neu.edu "cd R && sbatch config.slurm" | tee slurm_job_id.txt 
        slurm_job_id=$(grep -oP '\d+' slurm_job_id.txt) 
        echo "Slurm Job ID is $slurm_job_id"
        echo "slurm_job_id=$slurm_job_id" >> $GITHUB_ENV  

    - name: Monitor Slurm Job
      run: |
        ssh raina.ans@login-00.discovery.neu.edu "
          while squeue -j ${{ env.slurm_job_id }} | grep -q ${{ env.slurm_job_id }}; do
            echo 'Job Id : ${{ env.slurm_job_id }} is still running...'
            sleep 10
          done
          echo 'Job has completed.'
        "

    - name: Fetch Output
      run: |
        # scp raina.ans@login-00.discovery.neu.edu:/work/VitekLab/Projects/Benchmarking/job_output.txt job_output.txt
        # scp raina.ans@login-00.discovery.neu.edu:/work/VitekLab/Projects/Benchmarking/job_error.txt job_error.txt
        scp raina.ans@login-00.discovery.neu.edu:/home/raina.ans/R/job_output.txt job_output.txt
        scp raina.ans@login-00.discovery.neu.edu:/home/raina.ans/R/job_error.txt job_error.txt

    - name: Extract and Check FDR Values
      run: |
        # Extract the relevant lines containing FDR values and process them
        grep -A 5 "FPR  Accuracy    Recall       FDR" job_output.txt > results_section.txt
        if awk -v threshold= $FDR_THRESHOLD 'NR > 1 { if ($5 > threshold) { print "Error: FDR too high in row " NR " with FDR=" $5; exit 1 } }' results_section.txt; then
          echo "All FDR values are within acceptable range."
        fi

    - name: Upload Output as Artifact
      uses: actions/upload-artifact@v4
      with:
        name: benchmark-output
        path: |
          job_output.txt
          job_error.txt