Submitit

Submitit

Submitit is a lightweight tool for submitting Python functions for computation within a Slurm cluster. It basically wraps submission and provide access to results, logs and more. Slurm is an open source, fault-tolerant, and highly scalable cluster management and job scheduling system for large and small Linux clusters. Submitit allows to switch seamlessly between executing on Slurm or locally. Source code, issues and pull requests can be found here.

Currently, development is in progress for an OAR plugin, to facilitate the passage between OAR and Slurm based resource managers. Source code for an OAR plugin can be found here. Fort the first version, the supported and not supported parameters are listed in the table below. The not supported functionalities are the tasks notion of Slurm, the memory management of the job, the checkpointing, the job array and the asynchronous job supports.

Comparison with Dask-jobqueue

The key difference with Submitit is that Dask-jobqueue distributes the jobs to a pool of Dask workers, while Submitit jobs are directly jobs on the cluster. In that sense Submitit is a lower level interface than Dask-jobqueue and you get more direct control over your jobs, including individual stdout and stderr, and possibly checkpointing in case of preemption and timeout. On the other hand, you should avoid submitting multiple small tasks with Submitit, which would create many independent jobs and possibly overload the cluster, while you can do it without any problem through Dask-jobqueue.

Submitit installation

	Note
	Submitit should be installed in a folder accessible by both the frontend and the nodes. For example, by installing it in your homedir, we are guaranteed that Submitit is available by both the frontend and the nodes.

	Note
	Python 3.6+ is required to install Submitit

Using pip

pip can be used to install the stable release of submitit:

fontend.site:

pip install --user submitit

To use the last version including the OAR plugin, an installation from Source can be done:

fontend.site:

pip install --user git+https://gitlab.inria.fr/moyens-de-calcul/submitit.git@master#egg=submitit

It is recommended to install python dependencies via a virtual environment. To do so, before running the pip command:

fontend.site:

cd my_submitit_jobqueue_project

fontend.site:

python3 -m venv env

fontend.site:

source env/bin/activate

fontend.site:

pip install git+https://gitlab.inria.fr/moyens-de-calcul/submitit.git@master#egg=submitit

Using conda

conda can be used to install submitit from the conda-forge:

fontend.site:

conda install -c conda-forge submitit

To use the last version including the OAR plugin, you can create a conda environment file(e.g. "conda-env-submitit.yml") as:

name: submitit
dependencies:
  - pip:
    - git+https://gitlab.inria.fr/moyens-de-calcul/submitit.git@master#egg=submitit

and then install the last version of Submitit using this environment file

fontend.site:

conda env create --file conda-env-submitit.yml

fontend.site:

source activate submitit

Basic usage

Performing an addition with Submitit

Here is a Python script example which allows to execute an addition job on Slurm, OAR or locally.

import submitit

def add(a, b):
    return a + b

# logs are dumped in the folder
executor = submitit.AutoExecutor(folder="log_test")

job_addition = executor.submit(add, 5, 7)  # will compute add(5, 7)
output = job_addition.result()  # waits for completion and returns output
print('job_addition output: ', output)
assert output == 12

The example script can be launched on frontend as follow:

flille:

python3 this-script.py

The addition job will be computed on the cluster. For each job, in the working folder that you defined (e.g., folder="log_test"), you will find a stdout log file jobId_log.out, a stderr log file jobId_log.err, a submission batch file jobId_submission.sh, a task file jobId_submitted.pkl and an output file jobId_result.pkl.

Advanced usage

Parameters

Parameters for cluster can be setted by update_parameters(**kwargs).

The AutoExecutor shown in the basic usage example above is the common submission interface, for OAR/Slurm clusters and local jobs.

To use the cluster specific parameters with the AutoExecutor, they must be appended by the cluster name, e.g., slurm_partition="cpu_devel", oar_queue="default". These cluster specific options will be ignored on other clusters.

executor = submitit.AutoExecutor(folder="log_test")
executor.update_parameters(slurm_partition="cpu_devel", oar_queue="default")

	Note
	Cluster specific parameters win over common parameters.

E.g. if both oar_walltime and timeout_min are provided, then:

• oar_walltime is used on the OAR cluster
• timeout_min is used on other clusters

The cluster specific parameters can also be used with cluster specific Executors, without the cluster name prefixes, e.g., SlurmExecutor, OarExecutor.

executor = submitit.OarExecutor(folder="log_test")
executor.update_parameters(walltime="0:0:5", queue="default")

The following table recaps the parameters supported by AutoExecutor, OARExecutor and SlurmExecutor:

AutoExecutor	OARExecutor	SlurmExecutor	Description
timeout_min	walltime in hh:mm:ss	time	timeout in minutes
name	n	job_name	'submitit' by default
nodes	nodes	nodes	number of nodes in int
	oar_queue	slurm_partition	string
gpus_per_node	gpu	gpus_per_node or --gres=gpu:xx	number of gpu in int
stderr_to_stdout	not supported	stderr_to_stdout	boolean
tasks_per_node	not supported	ntasks_per_node	int
cpus_per_task	not supported	cpus_per_task	int
mem_gb	not supported	mem	string

Checkpointing with Submitit

Install scikit-learn (numpy, scipy required)