slurm-example.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "41853895-e5d7-4139-970b-34fdc2edb310",
   "metadata": {},
   "outputs": [],
   "source": [
    "import ipyparallel as ipp"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "748dca8f-25e3-473e-8dd3-1f09e0c12c76",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Using existing profile dir: '/data/profile'\n",
      "Starting 2 engines with <class 'ipyparallel.cluster.launcher.SlurmEngineSetLauncher'>\n",
      "Job submitted with job id: '14'\n",
      "100%|███████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 2/2 [00:08<00:00,  4.36s/engine]\n"
     ]
    }
   ],
   "source": [
    "cluster = ipp.Cluster(engine_launcher_class=\"slurm\", controller_ip=\"0.0.0.0\", profile_dir=\"/data/profile\")\n",
    "cluster.config.SlurmEngineSetLauncher.batch_template = \"\"\"\\\n",
    "#!/bin/sh\n",
    "#SBATCH --ntasks={n}\n",
    "#SBATCH --export=ALL\n",
    "#SBATCH --job-name=ipengine-{cluster_id}\n",
    "#SBATCH --mem=1gb\n",
    "#SBATCH --time=00:05:00\n",
    "#SBATCH --ntasks-per-node=1\n",
    "env | sort\n",
    "\n",
    "srun {program_and_args}\n",
    "\"\"\"\n",
    "\n",
    "cluster.start_cluster_sync(n=2)\n",
    "rc = cluster.connect_client_sync()\n",
    "rc.wait_for_engines(2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "e6ffeba7-a5f3-4fef-878e-3ff9d3c7f734",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "             JOBID PARTITION     NAME     USER ST       TIME  NODES NODELIST(REASON)\n",
      "                14    normal ipengine     root  R       0:06      2 c[1-2]\n"
     ]
    }
   ],
   "source": [
    "!squeue"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "11ea5231-bca2-47a0-aff6-0aea26d1ceea",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{0: 'c1', 1: 'c2'}"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import socket\n",
    "rc[:].apply_async(socket.gethostname).get_dict()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "fea4d80b-2953-4ad5-b599-5707698c6639",
   "metadata": {},
   "source": [
    "The rendered template:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "a0877530-f510-4c3b-b49a-7dce836b1c34",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "#!/bin/sh\n",
      "#SBATCH --ntasks=2\n",
      "#SBATCH --export=ALL\n",
      "#SBATCH --job-name=ipengine-1634581693-wslm\n",
      "#SBATCH --mem=1gb\n",
      "#SBATCH --time=00:05:00\n",
      "#SBATCH --ntasks-per-node=1\n",
      "env | sort\n",
      "\n",
      "srun /usr/bin/python3 -m ipyparallel.engine\n"
     ]
    }
   ],
   "source": [
    "!cat slurm_engine.sbatch"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.10"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}