Skip to content

Commit 6bb673d

Browse files
bclaremarbclaremar
authored
IDEs_cmd.rst reorder tabs jupyter
1 parent 0d963f1 commit 6bb673d

File tree

1 file changed

+114
-114
lines changed

1 file changed

+114
-114
lines changed

‎docs/day2/IDEs_cmd.rst‎

Lines changed: 114 additions & 114 deletions
Original file line numberDiff line numberDiff line change
@@ -69,6 +69,120 @@ Local notes
6969

7070
.. tabs::
7171

72+
.. tab:: NSC
73+
74+
**Through ThinLinc**
75+
76+
1. Login with ThinLinc (https://www.nsc.liu.se/support/graphics/)
77+
78+
- Download the client matching your local computer's OS and install it.
79+
- Start the ThinLinc client.
80+
- Change the “Server” setting to ``tetralith.nsc.liu.se``.
81+
- Change the “Name” setting to your Tetralith username (e.g x_abcde).
82+
- Enter your cluster Tetralith password in the “Password” box.
83+
- Press the “Connect” button.
84+
- If you connect for the first time, you will see the “The server’s host key is not cached …” dialog.
85+
86+
2. Load a JupyterLab module
87+
88+
- Open a terminal
89+
- This is an example for JupyterLab 4.2.0
90+
91+
.. code-block:: console
92+
93+
$ module load buildtool-easybuild/4.8.0-hpce082752a2 GCC/13.2.0 Python/3.11.5 SciPy-bundle/2023.11 JupyterLab/4.2.0
94+
95+
3. Start JupyterLab
96+
97+
- Type ``jupyter-lab`` in the terminal
98+
- It will show some text, including telling you to open a url in a browser (inside ThinLinc/on Tetralith). If you just wait, it will open a browser with Jupyter.
99+
100+
- It will look similar to this:
101+
102+
.. admonition:: Webbrowser view
103+
:class: dropdown
104+
105+
.. figure:: ../img/jupyter-thinlinc-nsc.png
106+
107+
108+
**On your own computer through SSH tunneling**
109+
110+
1. Either do a regular SSH or use ThinLinc to connect to tetralith (change to your own username):
111+
112+
``ssh x_abcde@tetralith.nsc.liu.se``
113+
114+
2. Change to your working directory
115+
116+
``cd <my-workdir>``
117+
118+
3. Load a module with JupyterLab in (here JupyterLab 4.2.0)
119+
120+
.. code-block:: console
121+
122+
$ module load buildtool-easybuild/4.8.0-hpce082752a2 GCC/13.2.0 Python/3.11.5 SciPy-bundle/2023.11 JupyterLab/4.2.0
123+
124+
4. Start jupyter with the no-browser flag
125+
126+
- ``jupyter-lab --no-browser``
127+
128+
- You get something that looks like this:
129+
130+
.. admonition:: Terminal view
131+
:class: dropdown
132+
133+
.. figure:: ../img/jupyter-no-browser-nsc.png
134+
135+
Where I have marked a line with relevant info. Note that the port will change.
136+
137+
5. Open a second terminal, on your home computer. Input this:
138+
139+
- ``ssh -N -L localhost:88XX:localhost:88XX x_abcde@tetralith1.nsc.liu.se``
140+
141+
where you change 88XX to the actual port you got, and the name to your username. In my example it would be:
142+
143+
- ``ssh -N -L localhost:8867:localhost:8867 x_birbr@tetralith1.nsc.liu.se``
144+
145+
.. admonition:: Terminal view
146+
:class: dropdown
147+
148+
.. figure:: ../img/local-ssh-to-nsc.png
149+
150+
6. Now grab the line that is similar to the one I marked in 4. and which has the same port as you used in 5.
151+
152+
- Input that line (url with token) in a browser on your local machine. You wil get something similar to this:
153+
154+
.. admonition:: Webbrowser view
155+
:class: dropdown
156+
157+
.. figure:: ../img/local-jupyter-lab.png
158+
159+
160+
.. tab:: PDC
161+
162+
You can interactively launch Jupyter Lab and Notebook on Dardel by following the steps as below. Hopefully the ThinLinc licenses are sufficient!
163+
164+
165+
1. Click on Applications -> PDC-Jupyter -> Jupyter Lab or Jupyter Notebook
166+
167+
.. admonition:: Desktop view
168+
:class: dropdown
169+
170+
.. figure:: ../img/pdc_start_jupyter.png
171+
172+
2. Configure your job parameters in the dialog box.
173+
174+
.. admonition:: GfxLauncher view
175+
:class: dropdown
176+
177+
.. figure:: ../img/pdc_jupyter_configure_job.png
178+
179+
3. Click Start, wait for the job to start and in few seconds a firefox browser will open with Jupyter Lab or Notebook session. If you close the firefox browser, you can connect to same Jupyter session again by clicking ‘Reconnect to Lab’.
180+
181+
.. admonition:: GfxLauncher view
182+
:class: dropdown
183+
184+
.. figure:: ../img/pdc_reconnect_to_jupyter.png
185+
72186
.. tab:: UPPMAX
73187

74188
**1. Login to a remote desktop**
@@ -396,120 +510,6 @@ Local notes
396510

397511
.. figure:: ../img/cosmos-on-demand-job-settings.png
398512

399-
.. tab:: NSC
400-
401-
**Through ThinLinc**
402-
403-
1. Login with ThinLinc (https://www.nsc.liu.se/support/graphics/)
404-
405-
- Download the client matching your local computer's OS and install it.
406-
- Start the ThinLinc client.
407-
- Change the “Server” setting to ``tetralith.nsc.liu.se``.
408-
- Change the “Name” setting to your Tetralith username (e.g x_abcde).
409-
- Enter your cluster Tetralith password in the “Password” box.
410-
- Press the “Connect” button.
411-
- If you connect for the first time, you will see the “The server’s host key is not cached …” dialog.
412-
413-
2. Load a JupyterLab module
414-
415-
- Open a terminal
416-
- This is an example for JupyterLab 4.2.0
417-
418-
.. code-block:: console
419-
420-
$ module load buildtool-easybuild/4.8.0-hpce082752a2 GCC/13.2.0 Python/3.11.5 SciPy-bundle/2023.11 JupyterLab/4.2.0
421-
422-
3. Start JupyterLab
423-
424-
- Type ``jupyter-lab`` in the terminal
425-
- It will show some text, including telling you to open a url in a browser (inside ThinLinc/on Tetralith). If you just wait, it will open a browser with Jupyter.
426-
427-
- It will look similar to this:
428-
429-
.. admonition:: Webbrowser view
430-
:class: dropdown
431-
432-
.. figure:: ../img/jupyter-thinlinc-nsc.png
433-
434-
435-
**On your own computer through SSH tunneling**
436-
437-
1. Either do a regular SSH or use ThinLinc to connect to tetralith (change to your own username):
438-
439-
``ssh x_abcde@tetralith.nsc.liu.se``
440-
441-
2. Change to your working directory
442-
443-
``cd <my-workdir>``
444-
445-
3. Load a module with JupyterLab in (here JupyterLab 4.2.0)
446-
447-
.. code-block:: console
448-
449-
$ module load buildtool-easybuild/4.8.0-hpce082752a2 GCC/13.2.0 Python/3.11.5 SciPy-bundle/2023.11 JupyterLab/4.2.0
450-
451-
4. Start jupyter with the no-browser flag
452-
453-
- ``jupyter-lab --no-browser``
454-
455-
- You get something that looks like this:
456-
457-
.. admonition:: Terminal view
458-
:class: dropdown
459-
460-
.. figure:: ../img/jupyter-no-browser-nsc.png
461-
462-
Where I have marked a line with relevant info. Note that the port will change.
463-
464-
5. Open a second terminal, on your home computer. Input this:
465-
466-
- ``ssh -N -L localhost:88XX:localhost:88XX x_abcde@tetralith1.nsc.liu.se``
467-
468-
where you change 88XX to the actual port you got, and the name to your username. In my example it would be:
469-
470-
- ``ssh -N -L localhost:8867:localhost:8867 x_birbr@tetralith1.nsc.liu.se``
471-
472-
.. admonition:: Terminal view
473-
:class: dropdown
474-
475-
.. figure:: ../img/local-ssh-to-nsc.png
476-
477-
6. Now grab the line that is similar to the one I marked in 4. and which has the same port as you used in 5.
478-
479-
- Input that line (url with token) in a browser on your local machine. You wil get something similar to this:
480-
481-
.. admonition:: Webbrowser view
482-
:class: dropdown
483-
484-
.. figure:: ../img/local-jupyter-lab.png
485-
486-
487-
.. tab:: PDC
488-
489-
You can interactively launch Jupyter Lab and Notebook on Dardel by following the steps as below. Hopefully the ThinLinc licenses are sufficient!
490-
491-
492-
1. Click on Applications -> PDC-Jupyter -> Jupyter Lab or Jupyter Notebook
493-
494-
.. admonition:: Desktop view
495-
:class: dropdown
496-
497-
.. figure:: ../img/pdc_start_jupyter.png
498-
499-
2. Configure your job parameters in the dialog box.
500-
501-
.. admonition:: GfxLauncher view
502-
:class: dropdown
503-
504-
.. figure:: ../img/pdc_jupyter_configure_job.png
505-
506-
3. Click Start, wait for the job to start and in few seconds a firefox browser will open with Jupyter Lab or Notebook session. If you close the firefox browser, you can connect to same Jupyter session again by clicking ‘Reconnect to Lab’.
507-
508-
.. admonition:: GfxLauncher view
509-
:class: dropdown
510-
511-
.. figure:: ../img/pdc_reconnect_to_jupyter.png
512-
513513
Spyder
514514
------
515515

0 commit comments

Comments
 (0)