diff --git a/distributed/ddp-tutorial-series/README.md b/distributed/ddp-tutorial-series/README.md
index d0ce17c00f..1424717f90 100644
--- a/distributed/ddp-tutorial-series/README.md
+++ b/distributed/ddp-tutorial-series/README.md
@@ -4,6 +4,14 @@ Code for the DDP tutorial series at https://pytorch.org/tutorials/beginner/ddp_s
 
 Each code file extends upon the previous one. The series starts with a non-distributed script that runs on a single GPU and incrementally updates to end with multinode training on a Slurm cluster.
 
+## Dependencies
+
+1. nccl
+   1. https://github.com/NVIDIA/nccl
+   2. https://github.com/NVIDIA/nccl-tests
+2. torch>=1.11.0
+
+
 ## Files
 * [single_gpu.py](single_gpu.py): Non-distributed training script
 
@@ -16,6 +24,32 @@ Each code file extends upon the previous one. The series starts with a non-distr
     * [slurm/config.yaml.template](slurm/config.yaml.template): configuration to set up an AWS cluster
     * [slurm/sbatch_run.sh](slurm/sbatch_run.sh): slurm script to launch the training job
 
+## Create Virtual Environment
+
+```shell
+$ python -m venv </path/to/new/virtual/environment>
+$ source </path/to/new/virtual/environment>/bin/activate
+```
+
+## Run commands
 
+* [single_gpu.py](single_gpu.py):
+```shell
+$ python single_gpu.py 50 10
+```
 
+* [multigpu.py](multigpu.py):
+
+```shell
+$ python multigpu.py 50 10
+```
+
+
+* [multigpu_torchrun.py](multigpu_torchrun.py):
+```shell
+$ torchrun --standalone --nproc_per_node=gpu multigpu_torchrun.py 50 10
+```
+
+* [multinode.py](multinode.py): DDP on multiple nodes using Torchrun (and optionally Slurm)
 
+  TODO
diff --git a/distributed/ddp-tutorial-series/datautils.py b/distributed/ddp-tutorial-series/datautils.py
index 22c364eaa0..52a0d6d07c 100644
--- a/distributed/ddp-tutorial-series/datautils.py
+++ b/distributed/ddp-tutorial-series/datautils.py
@@ -1,6 +1,7 @@
 import torch
 from torch.utils.data import Dataset
 
+
 class MyTrainDataset(Dataset):
     def __init__(self, size):
         self.size = size
@@ -10,4 +11,4 @@ def __len__(self):
         return self.size
     
     def __getitem__(self, index):
-        return self.data[index]
\ No newline at end of file
+        return self.data[index]
diff --git a/distributed/ddp-tutorial-series/multigpu.py b/distributed/ddp-tutorial-series/multigpu.py
index 029731b5d2..25864d72a0 100644
--- a/distributed/ddp-tutorial-series/multigpu.py
+++ b/distributed/ddp-tutorial-series/multigpu.py
@@ -2,12 +2,15 @@
 import torch.nn.functional as F
 from torch.utils.data import Dataset, DataLoader
 from datautils import MyTrainDataset
+from icecream import ic
 
+# --- Additional modules required for Distributed Training
 import torch.multiprocessing as mp
 from torch.utils.data.distributed import DistributedSampler
 from torch.nn.parallel import DistributedDataParallel as DDP
 from torch.distributed import init_process_group, destroy_process_group
 import os
+# ---
 
 
 def ddp_setup(rank, world_size):
@@ -46,7 +49,7 @@ def _run_batch(self, source, targets):
 
     def _run_epoch(self, epoch):
         b_sz = len(next(iter(self.train_data))[0])
-        print(f"[GPU{self.gpu_id}] Epoch {epoch} | Batchsize: {b_sz} | Steps: {len(self.train_data)}")
+        print(f"[GPU{self.gpu_id}] Epoch {epoch} | Batch size: {b_sz} | Steps: {len(self.train_data)}")
         self.train_data.sampler.set_epoch(epoch)
         for source, targets in self.train_data:
             source = source.to(self.gpu_id)
@@ -84,6 +87,8 @@ def prepare_dataloader(dataset: Dataset, batch_size: int):
 
 
 def main(rank: int, world_size: int, save_every: int, total_epochs: int, batch_size: int):
+    ic(rank, world_size)
+
     ddp_setup(rank, world_size)
     dataset, model, optimizer = load_train_objs()
     train_data = prepare_dataloader(dataset, batch_size)
diff --git a/distributed/ddp-tutorial-series/multigpu_torchrun.py b/distributed/ddp-tutorial-series/multigpu_torchrun.py
index 66d8187346..39a521ee39 100644
--- a/distributed/ddp-tutorial-series/multigpu_torchrun.py
+++ b/distributed/ddp-tutorial-series/multigpu_torchrun.py
@@ -3,17 +3,21 @@
 from torch.utils.data import Dataset, DataLoader
 from datautils import MyTrainDataset
 
-import torch.multiprocessing as mp
+# --- Additional modules required for Distributed Training
 from torch.utils.data.distributed import DistributedSampler
 from torch.nn.parallel import DistributedDataParallel as DDP
 from torch.distributed import init_process_group, destroy_process_group
 import os
+# ---
+
+from utils import print_nodes_info
 
 
 def ddp_setup():
     init_process_group(backend="nccl")
     torch.cuda.set_device(int(os.environ["LOCAL_RANK"]))
 
+
 class Trainer:
     def __init__(
         self,
@@ -52,7 +56,7 @@ def _run_batch(self, source, targets):
 
     def _run_epoch(self, epoch):
         b_sz = len(next(iter(self.train_data))[0])
-        print(f"[GPU{self.gpu_id}] Epoch {epoch} | Batchsize: {b_sz} | Steps: {len(self.train_data)}")
+        print(f"[GPU{self.gpu_id}] Epoch {epoch} | Batch size: {b_sz} | Steps: {len(self.train_data)}")
         self.train_data.sampler.set_epoch(epoch)
         for source, targets in self.train_data:
             source = source.to(self.gpu_id)
@@ -107,5 +111,9 @@ def main(save_every: int, total_epochs: int, batch_size: int, snapshot_path: str
     parser.add_argument('save_every', type=int, help='How often to save a snapshot')
     parser.add_argument('--batch_size', default=32, type=int, help='Input batch size on each device (default: 32)')
     args = parser.parse_args()
-    
+
+    # --- Print the environment variables
+    print_nodes_info()
+    # ---
+
     main(args.save_every, args.total_epochs, args.batch_size)
diff --git a/distributed/ddp-tutorial-series/multinode.py b/distributed/ddp-tutorial-series/multinode.py
index e80636bcc4..893989931b 100644
--- a/distributed/ddp-tutorial-series/multinode.py
+++ b/distributed/ddp-tutorial-series/multinode.py
@@ -2,18 +2,22 @@
 import torch.nn.functional as F
 from torch.utils.data import Dataset, DataLoader
 from datautils import MyTrainDataset
+from utils import print_nodes_info
 
+# --- Additional modules required for Distributed Training
 import torch.multiprocessing as mp
 from torch.utils.data.distributed import DistributedSampler
 from torch.nn.parallel import DistributedDataParallel as DDP
 from torch.distributed import init_process_group, destroy_process_group
 import os
+# ---
 
 
 def ddp_setup():
     init_process_group(backend="nccl")
     torch.cuda.set_device(int(os.environ["LOCAL_RANK"]))
 
+
 class Trainer:
     def __init__(
         self,
@@ -53,7 +57,7 @@ def _run_batch(self, source, targets):
 
     def _run_epoch(self, epoch):
         b_sz = len(next(iter(self.train_data))[0])
-        print(f"[GPU{self.global_rank}] Epoch {epoch} | Batchsize: {b_sz} | Steps: {len(self.train_data)}")
+        print(f"[GPU{self.global_rank}] Epoch {epoch} | Batch size: {b_sz} | Steps: {len(self.train_data)}")
         self.train_data.sampler.set_epoch(epoch)
         for source, targets in self.train_data:
             source = source.to(self.local_rank)
@@ -108,5 +112,9 @@ def main(save_every: int, total_epochs: int, batch_size: int, snapshot_path: str
     parser.add_argument('save_every', type=int, help='How often to save a snapshot')
     parser.add_argument('--batch_size', default=32, type=int, help='Input batch size on each device (default: 32)')
     args = parser.parse_args()
-    
+
+    # --- Print the environment variables
+    print_nodes_info()
+    # ---
+
     main(args.save_every, args.total_epochs, args.batch_size)
diff --git a/distributed/ddp-tutorial-series/requirements.txt b/distributed/ddp-tutorial-series/requirements.txt
index 9270a1d6ee..18ef26580e 100644
--- a/distributed/ddp-tutorial-series/requirements.txt
+++ b/distributed/ddp-tutorial-series/requirements.txt
@@ -1 +1,3 @@
-torch>=1.11.0
\ No newline at end of file
+torch>=1.11.0
+numpy
+icecream
\ No newline at end of file
diff --git a/distributed/ddp-tutorial-series/run_multigpu.sh b/distributed/ddp-tutorial-series/run_multigpu.sh
new file mode 100644
index 0000000000..77abc5930c
--- /dev/null
+++ b/distributed/ddp-tutorial-series/run_multigpu.sh
@@ -0,0 +1 @@
+python multigpu.py 50 10
\ No newline at end of file
diff --git a/distributed/ddp-tutorial-series/run_multigpu_torchrun.sh b/distributed/ddp-tutorial-series/run_multigpu_torchrun.sh
new file mode 100644
index 0000000000..7d4c594f6f
--- /dev/null
+++ b/distributed/ddp-tutorial-series/run_multigpu_torchrun.sh
@@ -0,0 +1 @@
+torchrun --standalone --nproc_per_node=gpu multigpu_torchrun.py 50 10
diff --git a/distributed/ddp-tutorial-series/run_single_gpu.sh b/distributed/ddp-tutorial-series/run_single_gpu.sh
new file mode 100644
index 0000000000..58e4e74816
--- /dev/null
+++ b/distributed/ddp-tutorial-series/run_single_gpu.sh
@@ -0,0 +1 @@
+python single_gpu.py 50 10
diff --git a/distributed/ddp-tutorial-series/single_gpu.py b/distributed/ddp-tutorial-series/single_gpu.py
index e91ab81cc1..d81c671f1c 100644
--- a/distributed/ddp-tutorial-series/single_gpu.py
+++ b/distributed/ddp-tutorial-series/single_gpu.py
@@ -11,7 +11,7 @@ def __init__(
         train_data: DataLoader,
         optimizer: torch.optim.Optimizer,
         gpu_id: int,
-        save_every: int, 
+        save_every: int,
     ) -> None:
         self.gpu_id = gpu_id
         self.model = model.to(gpu_id)
@@ -28,7 +28,7 @@ def _run_batch(self, source, targets):
 
     def _run_epoch(self, epoch):
         b_sz = len(next(iter(self.train_data))[0])
-        print(f"[GPU{self.gpu_id}] Epoch {epoch} | Batchsize: {b_sz} | Steps: {len(self.train_data)}")
+        print(f"[GPU{self.gpu_id}] Epoch {epoch} | Batch size: {b_sz} | Steps: {len(self.train_data)}")
         for source, targets in self.train_data:
             source = source.to(self.gpu_id)
             targets = targets.to(self.gpu_id)
diff --git a/distributed/ddp-tutorial-series/utils.py b/distributed/ddp-tutorial-series/utils.py
new file mode 100644
index 0000000000..d3ebf29e0a
--- /dev/null
+++ b/distributed/ddp-tutorial-series/utils.py
@@ -0,0 +1,10 @@
+import os
+
+from icecream import ic
+
+
+def print_nodes_info():
+    Node = os.environ['LOCAL_RANK']
+    ic(Node, os.environ['LOCAL_RANK'], os.environ['RANK'], os.environ['WORLD_SIZE'], os.environ['MASTER_ADDR'],
+       os.environ['MASTER_PORT'])
+    print()