Quick Start

Quick Start#

This guide will help you get started with srunx quickly.

Basic Job Submission#

Submit a simple Python script:

srunx submit python my_script.py

Submit with specific resources:

srunx submit python train.py --name ml_job --gpus-per-node 1 --nodes 2

Submit with conda environment:

srunx submit python process.py --conda ml_env --memory-per-node 32GB

Job Management#

Check job status:

srunx status <job_id>

List your jobs:

srunx list

Cancel a job:

srunx cancel <job_id>

Workflow Example#

Create a workflow YAML file (workflow.yaml):

name: ml_pipeline
jobs:
  - name: preprocess
    command: ["python", "preprocess.py"]
    resources:
      nodes: 1

  - name: train
    command: ["python", "train.py"]
    depends_on: [preprocess]
    resources:
      gpus_per_node: 1
      memory_per_node: "32GB"
      time_limit: "4:00:00"
    environment:
      conda: ml_env

  - name: evaluate
    command: ["python", "evaluate.py"]
    depends_on: [train]

Run the workflow:

srunx flow run workflow.yaml

Validate a workflow:

srunx flow validate workflow.yaml

Environment Setup#

srunx supports multiple environment types:

Conda Environment#

srunx submit python script.py --conda my_env

Python Virtual Environment#

srunx submit python script.py --venv /path/to/venv

Container (Pyxis)#

srunx submit python script.py --container /path/to/container.sqsh

Apptainer / Singularity Container#

srunx submit python script.py \
  --container "runtime=apptainer,image=/path/to/image.sif,nv=true"

Or specify the runtime separately:

srunx submit python script.py \
  --container /path/to/image.sif \
  --container-runtime apptainer

Conda Inside a Container#

Containers can be combined with conda or venv:

srunx submit python script.py \
  --container "runtime=apptainer,image=pytorch.sif,nv=true,bind=/data:/data" \
  --conda ml_env

Next Steps#

Contents