Pre-requisites for Connecting to ARC Cluster
Access to the ARC Cluster requires very important pre-requisites. Without these, you will not be able to access the ARC Cluster. First,
- An account is required to access the ARC cluster. Your account will be created prior to the workshop.
- You must be connected to the UC campus network to access the cluster. If you are accessing from an off-campus network, you need to connect with UC's VPN.
- An ssh (secure shell) client is required to login to the ARC Cluster. Most of the Operating Systems will have a built-in ssh client. You can type 'ssh' in the command terminal to check for the availability.
- Some older versions of Windows don't have an ssh client. In this case, please download PuTTY, a free and open source ssh application. Click here to download PuTTY.
- If you don't have a native ssh client, click here for instructions on how to connect to ARC Cluster using PuTTY.
Connecting to ARC Cluster
Every first-time user has to login to the ARC Cluster through the command terminal using an ssh client. This helps to create a 'Home Directory' on your account, where you can upload and download the files. Please go through the following instructions to establish a connection to the ARC Cluster.
- Use a command terminal to login to the cluster by typing - ssh username@arcc2.uc.edu. Replace "username" with your UC central login 6+2 ID.
- First time access from any PC will have to authenticate the connection. Once you type "ssh username@arcc2.uc.edu" and press enter, a notification, “Authenticity can’t be established”, will be returned. Type “yes” and press "enter" to authenticate. See the example below where “username” is 'yellugki'.
- Type “ssh username@arcc2.uc.edu” again and enter the password (same as the UC central login password). You are now connected to the ARC cluster!
- If you are using PuTTY, type 'arcc.uc.edu' under Host Name and click 'Open'. This will open an ssh client. Now, enter your username, the 6+2 ID, under 'login as'.
Cloning the Repositories
All the Python workshop material is available on GitHub. 'git clone' command is used to clone (download) any repository from GitHub. Provide the URL (web address) after typing 'git clone' in the terminal.
- Now, clone (download) the respective workshop's repository to your account by typing the following commands in the terminal.
- git clone https://github.com/The-CEAS-Library/Foundations-of-Python-1
- git clone https://github.com/The-CEAS-Library/Foundations-of-Python-2
- git clone https://github.com/The-CEAS-Library/Foundations-of-Python-3
- git clone https://github.com/The-CEAS-Library/Data_Manipulation_with_Python
- git clone https://github.com/The-CEAS-Library/Data_Visualization_with_Python
- Once you clone a repository, all the files of that repository will be available on your account.
Accessing OnDemand Web Portal
OnDemand web portal is used to edit, upload/download files, and to create sessions on interactive apps like Jupyter Notebook, a web based application execute Python code.
- After you login to the ARC Cluster through the terminal, type the URL - https://ondemand.arc.uc.edu/ in a web-browser to login to the OnDemand web portal. Use your 6+2 ID for the username and the same UC central login password.
- You can also clone a repository through the OnDemand portal's shell access. Click the 'Clusters' tab first and then the 'ARCC Shell Access' to launch the shell (terminal).
- The shell's interface on the OnDemand portal is similar to the command terminal's interface (Please look at the initial instructions of this page). Now, you can clone any repository through the shell.
- Click the “Interactive Apps” tab and then select 'Jupyter Notebook'.
- Once you launch a session on OnDemand, it will be queued before the session starts. As the session starts, you can click on 'Connect to Jupyter' to launch Jupyter.
- 'Connect to Jupyter' will take you to the home directory on Jupyter. All the repositories that are cloned from GitHub will be displayed on this home directory.
- Click the respective workshop's folder, for example here - 'Introduction_to_Python_Part1'. There will be multiple files inside each workshop folder. All Jupyter notebooks will have respective workshop's name, and a '.ipynb' file extension. Click the Jupyter notebook to open the workshop content.
