TACC Instructions
User Manual:
Open the PDF directly: View PDF 
.
Page Count: 1
Data 1401 Labs
Instructions for using TACC Visualization Portal
These instructions will lead you through starting your Jupyter sessions on TACC’s Visualization Portal. In lab 1 you
registered for an account and e-mailed your account information to Dr. Farbin. You should be now added to the resource
allocation for this lab, thereby giving you access to the TACC Visualization Portal.
● Login:
● Navigate your browser to vis.tacc.utexas.edu, and log in with your TACC username and password.
● Once logged in, you should be on the “Jobs” tab (top left). If not, switch to the “Jobs” tab.
● If you have a job already running, simply use it and skip starting a new job.
● Start a new job:
● In the “Start a Job” box, select “Stampede2”.
● Your project should already selected as “DATA-1401-Introducti”.
● Select “iPython/Jupyter Notebook” as your session type.
● Select “development”, “normal”, or “large” as the queue. The “development” queue will give you a 2 hour
work session. You may select the others if you will need more time.
● Press “Start Job”.
● The system will take about a minute to submit and start your job. Just wait unit the page refreshes.
● Connect to your Jupyter session:
● Press the green “Open in Browser” button. A new tab should open with your own Jupyter server running
and showing the contents of your home directory.
● You may now work on your lab.
● Note that while you may choose to have your Jupyter session run when you are not working, you will be using
allocated time on the system. So if you will need to walk away or pause working for a while and are not explicitly
waiting for a long computation to finish, please save your work and end your session.
● Once done or ready to end your session:
● Come back to the main Jupyter page by either going back to the original Jupyter tab or clicking on the
Jupyter name/icon on top of page.
● Press the logout button on top right of the page.
● Go back to the TACC Visualization Portal Jobs tab and press “Terminate Jupyter”.
●
