Streamlined slurm script and removed references to redis_password (#10827)

Co-authored-by: Ralph Kube <ralph.kube@uit.not>
Co-authored-by: Richard Liaw <rliaw@berkeley.edu>
Resolved  Conflicts:
	doc/source/cluster/slurm.rst

doc/source/cluster/slurm.rst

@@ -21,57 +21,48 @@ Here are some community-contributed templates for using SLURM with Ray:
 
 .. _`Template script`: https://gist.github.com/pengzhenghao/b348db1075101a9b986c4cdfea13dcd6
 
+
 Starter SLURM script
 --------------------
 
 .. code-block:: bash
 
   #!/bin/bash
-
   #SBATCH --job-name=test
   #SBATCH --cpus-per-task=5
   #SBATCH --mem-per-cpu=1GB
-  #SBATCH --nodes=3
-  #SBATCH --tasks-per-node 1
+  #SBATCH --nodes=4
+  #SBATCH --tasks-per-node=1
+  #SBATCH --time=00:30:00
+  #SBATCH --reservation=test
 
-  worker_num=2 # Must be one less that the total number of nodes
+  let "worker_num=(${SLURM_NTASKS} - 1)"
 
-  # module load Langs/Python/3.6.4 # This will vary depending on your environment
-  # source venv/bin/activate
-
-  nodes=$(scontrol show hostnames $SLURM_JOB_NODELIST) # Getting the node names
-  nodes_array=( $nodes )
-
-  node1=${nodes_array[0]}
-
-  ip_prefix=$(srun --nodes=1 --ntasks=1 -w $node1 hostname --ip-address) # Making address
-  suffix=':6379'
-  ip_head=$ip_prefix$suffix
-  redis_password=$(uuidgen)
+  # Define the total number of CPU cores available to ray
+  let "total_cores=${worker_num} * ${SLURM_CPUS_PER_TASK}"
 
+  suffix='6379'
+  ip_head=`hostname`:$suffix
   export ip_head # Exporting for latter access by trainer.py
 
-  srun --nodes=1 --ntasks=1 -w $node1 ray start --block --head --redis-port=6379 --redis-password=$redis_password & # Starting the head
+  # Start the ray head node on the node that executes this script by specifying --nodes=1 and --nodelist=`hostname`
+  # We are using 1 task on this node and 5 CPUs (Threads). Have the dashboard listen to 0.0.0.0 to bind it to all
+  # network interfaces. This allows to access the dashboard through port-forwarding:
+  # Let's say the hostname=cluster-node-500 To view the dashboard on localhost:8265, set up an ssh-tunnel like this: (assuming the firewall allows it)
+  # $  ssh -N -f -L 8265:cluster-node-500:8265 user@big-cluster
+  srun --nodes=1 --ntasks=1 --cpus-per-task=${SLURM_CPUS_PER_TASK} --nodelist=`hostname` ray start --head --block --dashboard-host 0.0.0.0 --port=6379 --num-cpus ${SLURM_CPUS_PER_TASK} &
   sleep 5
   # Make sure the head successfully starts before any worker does, otherwise
   # the worker will not be able to connect to redis. In case of longer delay,
   # adjust the sleeptime above to ensure proper order.
-
-  # Now we execute worker_num worker nodes on all nodes in the allocation except hostname by
+  
+  # Now we execute worker_num worker nodes on all nodes in the allocation except hostname by 
   # specifying --nodes=${worker_num} and --exclude=`hostname`. Use 1 task per node, so worker_num tasks in total
   # (--ntasks=${worker_num}) and 5 CPUs per task (--cps-per-task=${SLURM_CPUS_PER_TASK}).
   srun --nodes=${worker_num} --ntasks=${worker_num} --cpus-per-task=${SLURM_CPUS_PER_TASK} --exclude=`hostname` ray start --address $ip_head --block --num-cpus ${SLURM_CPUS_PER_TASK} &
   sleep 5
 
-  for ((  i=1; i<=$worker_num; i++ ))
-  do
-    node2=${nodes_array[$i]}
-    srun --nodes=1 --ntasks=1 -w $node2 ray start --block --address=$ip_head --redis-password=$redis_password & # Starting the workers
-    # Flag --block will keep ray process alive on each compute node.
-    sleep 5
-  done
-
-  python -u trainer.py $redis_password 15 # Pass the total number of allocated CPUs
+  python -u trainer.py ${total_cores} # Pass the total number of allocated CPUs
 
 .. code-block:: python
 
@@ -82,10 +73,9 @@ Starter SLURM script
   import time
   import ray
 
-  redis_password = sys.argv[1]
-  num_cpus = int(sys.argv[2])
+  num_cpus = int(sys.argv[1])
 
-  ray.init(address=os.environ["ip_head"], _redis_password=redis_password)
+  ray.init(address=os.environ["ip_head"])
 
   print("Nodes in the Ray cluster:")
   print(ray.nodes())