Slurm Commands

Here is a list of commands used in slurm job scheduler testing.

• Check slurm node status:

  $ sinfo -N -l
  

• Transfer node state from drained to idle:

  $ scontrol update NodeName=c01 State=DOWN Reason="undraining"
  $ scontrol update NodeName=c0$i State=RESUME
  

• Check job status

  squeue
  

Sample Job Scripts

These scripts are copied from CECI Website.

Serial Job

test.sh

#!/bin/bash
#
#SBATCH --job-name=test
#SBATCH --output=res.txt
#
#SBATCH --ntasks=1
#SBATCH --time=10:00
#SBATCH --mem-per-cpu=100

srun hostname
srun sleep 60

Job submission

sbatch test.sh

MPI Job

mpitest.sh

#!/bin/bash
#
#SBATCH --job-name=test_mpi
#SBATCH --output=res_mpi.txt
#
#SBATCH --ntasks=6
#SBATCH --time=10:00
#SBATCH --mem-per-cpu=100

/shared/openmpi/bin/mpirun mpitest.out

c source code

mpitest.c

This sample MPI source code is copied from Wikipedia with minor modification.

/*
  "Hello World" MPI Test Program
*/
#include <assert.h>
#include <stdio.h>
#include <string.h>
#include <mpi.h>

int main(int argc, char **argv)
{
    char buf[256];
    int my_rank, num_procs;

    /* Initialize the infrastructure necessary for communication */
    MPI_Init(&argc, &argv);

    /* Identify this process */
    MPI_Comm_rank(MPI_COMM_WORLD, &my_rank);

    /* Find out how many total processes are active */
    MPI_Comm_size(MPI_COMM_WORLD, &num_procs);

    // Get the name of the processor
    char processor_name[MPI_MAX_PROCESSOR_NAME];
    int name_len;
    MPI_Get_processor_name(processor_name, &name_len);

    /* Until this point, all programs have been doing exactly the same.
       Here, we check the rank to distinguish the roles of the programs */
    if (my_rank == 0) {
        int other_rank;
        printf("We have %i processes.\n", num_procs);

        /* Send messages to all other processes */
        for (other_rank = 1; other_rank < num_procs; other_rank++)
        {
            sprintf(buf, "Hello %i!", other_rank);
            MPI_Send(buf, sizeof(buf), MPI_CHAR, other_rank,
                     0, MPI_COMM_WORLD);
        }

        /* Receive messages from all other process */
        for (other_rank = 1; other_rank < num_procs; other_rank++)
        {
            MPI_Recv(buf, sizeof(buf), MPI_CHAR, other_rank,
                     0, MPI_COMM_WORLD, MPI_STATUS_IGNORE);
            printf("%s\n", buf);
        }

    } else {

        /* Receive message from process #0 */
        MPI_Recv(buf, sizeof(buf), MPI_CHAR, 0,
                 0, MPI_COMM_WORLD, MPI_STATUS_IGNORE);
        assert(memcmp(buf, "Hello ", 6) == 0),

        /* Send message to process #0 */
        sprintf(buf, "Process %i reporting for duty from processor %s.", my_rank, processor_name);
        MPI_Send(buf, sizeof(buf), MPI_CHAR, 0,
                 0, MPI_COMM_WORLD);

    }

    /* Tear down the communication infrastructure */
    MPI_Finalize();
    return 0;
}

Compile source code

mpicc mpitest.c -o mpitest.out

Job submission

sbatch mpitest.sh