Job Submission

NU HPC clusters use SLURM workload manager to schedule, distribute, and execute user jobs. SLURM (the name comes from Simple Linux Utility for Resource Management) is free and open-source software used by many, if not most, large HPC facilities throughout the world. Thus, if you happen to have run research calculations at some HPC facility elsewhere, it should be rather easy for you to migrate your jobs and start using NU HPC clusters. On a login node, users arrange their data, write batch scripts, and submit their jobs to the execution queue. The submitted jobs are then put in a pending state until the requested system resources are available and allocated. SLURM will schedule each job to run in the queue according to a predetermined site policy designated to balance competing user needs and to maximize efficient use of the cluster resources.

Absolutely all computational tasks on Shabyt (apart from compiling and very short test runs that use one or just few cores) are supposed to be executed using SLURM workload manager that distributes them across the system in an optimal way. It is extremely important that users do not abuse the management/login node (ln01 in Shabyt) where they log in and do not run long heavy calculations on it interactively or in the background. The function of the management node is to let users compile binaries, transfer data, manage files, prepare input for calculations, and submit jobs. The management node is not a workhorse for heavy calculations.

For a comprehensive guide on SLURM you can refer to its website. A short printable cheat sheet of some useful SLURM commands and parameters is available in this summary. Below we will dive into explaining how jobs should be submitted for execution in NU HPC systems and provide some basic examples.

Partitions, QoS, and job limits

Each job’s position and progress in the queue is determined through the fairshare algorithm, which depends on a number of factors (e.g. size of the job, time requirement, job queuing time, partition, QoS, etc). For the information on the available partitions, Quality of Service (QoS), maximum job durations, and maximum number of simultaneously running jobs and CPU cores used, please refer to the corresponding sections on page Policies. Please note that the limits are subject to change.

It is important to keep in mind that your jobs cannot request the duration that exceed the time limit set by our policies. Jobs that request execution times exceeding the limit may still be sent to the queue, but they will stay queued (i.e. will be in a pending state) forever until you change the time requested. Likewise, requesting more RAM for your job than what is physically available will result in your submitted job stay in a pending state forever. One cannot simultaneously use more CPU cores than what is allowed for a single user that belongs to a specific QoS category. For example, if the total CPU core limit is 256 cores for a user and this user currently has four 64-core jobs running, then any new submitted jobs will be placed in a queue in stay in a pending state. They will not start running even if there are resources available to execute them. They will be pending until one or all of the four running jobs finish.

Job submission

Jobs can be submitted for execution using a “batch” file. The batch file is essentially a Linux shell script (typically a bash script, but using other shells, e.g. tcsh, is possible as well) that in addition to the actual user commands contains a preamble (or header) written in a special format. This header, all lines of which begin with the keyword #SBATCH, contains the information about the resources requested for the job, user information, job name, etc. While bash and other Linux shells treat these lines beginning with #SBATCH as comments, these are not comments for SLURM. SLURM reads them when you submit a job for execution, interprets them, and acts accordingly. Note that if you change the format of those lines just slightly, e.g. if instead they begin with # SBATCH or ##SBATCH then SLURM no longer reads them and assumes they are comments. This is convenient for making SLURM omit some lines without actually deleting them in your script.

Most common and useful SLURM commands

Common and Useful SLURM Commands

Command	Description
sbatch script.sh	Submit a job using the specified script (`script.sh`).
srun	Run a command or script directly under SLURM control, often used within scripts for MPI tasks.
scancel job_id	Cancel a running or queued job by specifying its job ID.
sacct	View accounting information for completed jobs.
squeue	View the status of jobs in the queue, including their state (running, pending, etc.).
sinfo	Display information about the SLURM cluster, including nodes, partitions, and their status.
scontrol show job job_id	Show detailed information about a specific job.
salloc	Allocate resources interactively, useful for debugging or interactive sessions.
sstat -j job_id	Display runtime statistics for a running job (e.g., CPU and memory usage).
sprio	Display job priorities in the queue.
smap	Display a graphical view of job allocations and node usage (requires smap installation).
sshare	Display a user's SLURM share information in a multi-user, shared-resource environment.
sbatch --test-only script.sh	Test the SLURM script for errors without submitting a job.
seff job_id	Show job efficiency information (if available), useful for analyzing resource usage efficiency.
squeue --start	Estimate the start time for queued jobs.

Running Serial / Single Threaded Jobs

First we are going to create basic python script called myscript.py

a = 10

for i in range(a);
  print('Hello World')

Serial or single CPU core jobs are those jobs that can only make use of one CPU on a node.

## Shebang
#!/bin/bash

## Resource request
#SBATCH --job-name=Test_Serial
#SBATCH --ntasks=1
#SBATCH --output=stdout%j.out
#SBATCH --error=stderr%j.out

## Bash command
python3 ./myscript.py
homedatectl

Distributed Memory Parallelism (MPI) Job

Message Passing Interface (MPI) is a standardized and portable message-passing standard designed to allow for execution of programs using CPUs on multiple nodes where CPUs across nodes communicate over the network. The MPI standard defines the syntax and semantics of library routines that are useful to a wide range of users writing portable message-passing programs in C, C++, and Fortran. Intel MPI and OpenMPI are available in Shabyt system and SLURM jobs may make use of either MPI implementations.

Requesting for multiple nodes and /or loading any MPI modules may not necessarily make your code faster, your code must be MPI aware to use MPI. Even though running a non-MPI code with mpirun might possibly succeed, you will most likely have every core assigned to your job running the exact computation, duplicating each others work, and wasting resources.

The version of the MPI commands you run must match the version of the MPI library used in compiling your code, or your job is likely to fail. And the version of the MPI daemons started on all the nodes for your job must also match. For example, an MPI program compiled with Intel MPI compilers should be executed using Intel MPI runtime instead of Open MPI runtime.

#!/bin/bash
#SBATCH --job-name=Test_MPI
#SBATCH --nodes=2
#SBATCH --ntasks=256
#SBATCH --ntasks-per-node=128
#SBATCH --time=0-0:30:00
#SBATCH --mem=32G
#SBATCH --partition=CPU

pwd; hostname; date
NP=${SLURM_NTASKS}
module load iimpi/2022b
mpirun -np ${NP} ./my_mpi_program <options>

GPU Job

#!/bin/bash
#SBATCH --job-name=gputest
#SBATCH --output=gpu.test.out
#SBATCH --error=gpu.test.err
#SBATCH --mail-type=ALL
#SBATCH --mail-user=email@nu.edu.kz
#SBATCH --nodes=1
#SBATCH --ntasks=8
#SBATCH --cpus-per-task=1
#SBATCH --ntasks-per-node=8
#SBATCH --distribution=cyclic:cyclic
#SBATCH --mem-per-cpu=7000mb
#SBATCH --partition=NVIDIA
#SBATCH --gpus=a100:4
#SBATCH --time=00:30:00

module purge
module load cuda/11.4.1 intel/2023b OpenMPI/4.0.5-GCC-9.3.0

SLURM Job Options

A SLURM script includes a list of SLURM job options at the top of the file, where each line starts with #SBATCH followed by option name to value pairs to tell the job scheduler the resources that a job requests.

Long Option	Short Option	Default value	Description
--job-name	-J	file name of job script	User defined name to identify a job
--time	-t	48:00:00	Specify a limit on the maximum execution time (walltime) for the job (D-HH:MM:SS) . For example, -t 1- is one day, -t 6:00:00 is 6 hours
--nodes	-N		Total number of node(s)
--ntasks	-n	1	Number of tasks (MPI workers)
--ntasks-per-node			Number of tasks per node
--cpus-per-task	-c	1	Number of CPUs required per task
--mem			Amount of memory allocated per node. Different units can be specified using the suffix [K|M|G|T]
--mem-per-cpu			Amount of memory allocated per cpu per code (For multicore jobs). Different units can be specified using the suffix [K|M|G|T]
--constraint	-C		Nodes with requested features. Multiple constraints may be specified with AND, OR, Matching OR. For example, --constraint="CPU_MNF:AMD", --constraint="CPU_MNF:INTEL&CPU_GEN:CLX"
--exclude	-x		Explicitly exclude certain nodes from the resources granted to the job.  For example, --exclude=cn[1-3]