Accessing JupyterHub (jupyter.cs) data
Unlike many JupyterHub deployments, your data is yours and have many different ways to access it. Thus, we don’t just have jupyter.cs, but a whole constellation of ways to access and do your work, depending on what suits you best for each part.
Your data (and as an instructor, your course’s data) can be accessed many ways:
On jupyter.cs.
Via network drive on your own computer as local files.
On Aalto shell servers (such as kosh.aalto.fi).
On other department/university workstations.
On Paniikki and Aalto computers
On Paniikki, and the Aalto servers kosh.aalto.fi
,
lyta.aalto.fi
, brute.aalto.fi
, and force.aalto.fi
(and
possibly more), the JupyterHub is available automatically. You can,
for example, use the Paniikki GPUs.
Data is available within the paths /m/jhnas/jupyter
. The path on
Linux servers is also available on the hub, if you want to write
portable files.
Name |
Path on hub |
Path on Linux servers |
---|---|---|
personal notebooks |
|
|
course data |
|
|
course instructor files |
|
|
shared data |
|
|
Variable seen above |
Meaning |
---|---|
|
Your Aalto username |
|
The two numbers you see in |
|
The short name of your course. |
You can change directly to your notebook directory by using cd
/m/jhnas/jupyter/${HOME%/unix}
.
You can link it to your home directory so that it’s easily
available. In a terminal, run /m/jhnas/jupyter/u/makedir.sh
and you
will automatically get a link from ~/jupyter
in your home
directory to your user data.
Permission denied? Run kinit
in the shell - this authenticates
yourself to the Aalto server and is required for secure access. If
you log in with ssh keys, you may need to do this.
Remote access via network drive
Basic info
Name |
Network drive path |
---|---|
personal notebooks |
|
course data |
|
course instructor files |
|
shared data |
|
You can do a SMB mount, which makes the data available as a network drive. You will have the same copy of the data as on the hub - actually, same data, so edits immediately take effect on both places, just like your home directory. You must be on an Aalto network, which for students practically means you must be connected to the Aalto VPN (see vpn instructions) or use an Aalto computer. The “aalto” wifi network does not work unless you have an Aalto computer.
Linux: use “Connect to Server” from the file browser. The path is
smb://jhnas.org.aalto.fi/$username
. You may need to useAALTO\username
as your username. If there is separate “domain” option, useAALTO
for domain and just your username for the username.Mac: same path as Linux above, “Connect to Server”. Use
AALTO\your_username
as the username.Windows:
\\jhnas.org.aalto.fi\$username
, and use usernameAALTO\your_username
. Windows sometimes caches the username/password for a long time, so if it does not work try rebooting.
You can also access course data and shared data by using
jhnas.org.aalto.fi/course/
or jhnas.org.aalto.fi/shareddata/
.
See also
Mounting network drives in Windows is the same instructions, but for Aalto home directories. Anything there should apply here, too.
Using GPUs
One problem with our JupyterHub so far is that we don’t have GPUs available. But, because our data is available to other computers, you can use the Paniikki: Computer Lab For Students GPUs (quite good ones) to get all the power you need. To do this, you just need to access the Jupyter data on these classroom computers.
Terminal: First, start a terminal. You can navigate to your data
following the instructions above: cd
/m/jhnas/jupyter/${HOME%/unix}
. From there, navigate to the right
directories and do what is needed.
File browser: Navigate to the path
/m/jhnas/jupyter/u/$nn/$username
, where $nn
is the two numbers
you see when you do echo $HOME
in a terminal. To open a terminal
from a location, right click and select “Open in Terminal”.
Now that you have the terminal and the data, you can do whatever you want with it. Presumably, you will start Jupyter here - but first you want to make the right software available. (Please don’t go installing large amounts of software in the Jupyter data directories - they are for notebooks and small-medium data.)