using jupyter kernel on vscode added

docs/VSCode.md

@@ -27,19 +27,16 @@ You can sync all your local VSCode extensions to the remote server after you are
 
 ### 2.3. Selecting Kernels
 
-First you need to start the correct module on Rackham that you contains the interpret you want on your VSCode. For example in case you need ML packages and python interpreter, do `module load python_ML_packages` on Rackham. Check the file path for python interpreter by checking `which python` and copy this path. In case of jupyter server, you can check `jupyter kernelspec list`.
-Go to Command Palette `Ctrl+Shift+P` or `F1` on your local VSCode. Search for "interpreter" for python or jupyter server, for example. Then paste the path of your interpreter/kernel. 
-
-`venv` or `conda` environments are also visible on VSCode when you select interpreter/kernel for python or jupyter server. For jupyter, you need to start the server first, check Point 4.
-
-## 3. Port forwarding to use Snowy gpus
-
+Request allocation in either Rackham or Snowy compute node depending on your need, for that use `interactive` slurm command. Load the correct module on Rackham/Snowy that you contains the interpret you want on your VSCode. For example in case you need ML packages and python interpreter, do `module load python_ML_packages`. Check the file path for python interpreter by checking `which python` and copy this path. Go to Command Palette `Ctrl+Shift+P` or `F1` on your local VSCode. Search for "interpreter" for python, then paste the path of your interpreter/kernel. 
 
+`venv` or `conda` environments are also visible on VSCode when you select interpreter/kernel for python or jupyter server. For jupyter, you need to start the server first, check Point 3.
 
 ## 4. Working with jupyter server on Rackham and snowy
 
 ### Rackham:
 
+Module load jupyter packages either from `module load python` or `module load python_ML_packages` as per your needs. For heavy compute and longer running jupyter server, allocate a Rackham compute node instead of using login node. Either request for rackham compute node by using, for example, `interactive -A uppmax202x-x-xx -p node -N 1 -t 2:00:00` or move to the next step to run jupyter on login node itself. Start the jupyter server `jupyter notebook --ip 0.0.0.0 --no-browser`. Copy the jupyter server URL which goes somethig like `http://r52.uppmax.uu.se:8888/tree?token=xxx`, click on **Select Kernel** on VSCode and select **Existing Jupyter Server**. Past the URL here and confirm your chioce. 
+
 ### Snowy:
 
-Start an interactive session with GPU allocation on Snowy `interactive -A uppmax202x-x-xx -p node -N 1 -t 02:00:00 --gres=gpu:1 -M snowy`. Module load the jupyter packages `module load python_ML_packages` and start the jupyter server `jupyter notebook --ip 0.0.0.0 --no-browser`. This should start a jupyter server on Snowy compute node with one T4 GPU. Copy the URL of the running jupyter server which goes something like `http://s193.uppmax.uu.se:8888/tree?token=xxx` and paste it in the jupyter kernel path on your local VSCode.
+Start an interactive session with GPU allocation on Snowy `interactive -A uppmax202x-x-xx -p node -N 1 -t 02:00:00 --gres=gpu:1 -M snowy`. Module load the jupyter packages `module load python_ML_packages` and start the jupyter server `jupyter notebook --ip 0.0.0.0 --no-browser`. This should start a jupyter server on Snowy compute node with one T4 GPU. Copy the URL of the running jupyter server which goes something like `http://s193.uppmax.uu.se:8888/tree?token=xxx` and paste it in the jupyter kernel path on your local VSCode. The application will automatically perform port forwarding to Rackham, which already is listening to Snowy compute nodes over certain ports.