Once your account has been approved, you will need to attend the following training sessions to continue using the ACCRE cluster. ACCRE requires that all new users complete the training within two months of receiving their accounts.
In accordance with Vanderbilt’s Return to Campus plan we are holding our ACCRE training sessions remotely using Zoom meetings. We recommend that if you have not used Zoom meetings before you attempt to connect 15 minutes early in order to ensure that the software installs correctly and that your audio connection works properly. If you are unable to connect to the remote meeting, please feel free to sign up for a future training session.
Please fill out a help desk ticket if you are unavailable at the time listed (for instance, if you have a class). We will provide you with a version of the class on Brightspace that you can take at your convenience.
Intro to Unix
This class gives an overview of Unix/Linux operating systems. You may skip this training session if you are already familiar with Linux or Unix.
Register (Tuesdays at 1:30pm or 2:30pm, and by appointment)
- Command line execution and shortcuts
- Getting help with commands
- Files and directories, including access rights
- Your Environment
- Text Editors
- Input/output redirection
- Pipelines and filters
Intro to the ACCRE Cluster
This class gives an overview of the ACCRE cluster, including basic commands. It is required for all new users.
Register (Wednesdays at 1:30pm or 2:30pm, and by appointment)
- Logging on to the cluster
- Transferring files to and from the cluster
- Setting your environment (Lmod)
- Scheduler basics
- Cluster etiquette
- Available storage
Intro to SLURM
This class gives an overview of SLURM, which is used to schedule computing jobs on the ACCRE cluster. It is required for all new users.
Please review the One Million Digits of Pi tutorial before attending this session.
Register (Thursdays at 1:30pm or 2:30pm, and by appointment)
- Components of the scheduler (SLURM) and their functions
- Submitting a job for execution
- Changing scheduler options after a job has been submitted
- Checking the status of a submitted job
- Checking your account’s status (fairshare)
- Using information about the overall cluster status (node availability) to optimize the timing of your job’s execution
Previously offered training sessions
Here are slide decks for training sessions that were previously offered by ACCRE. They are included here for reference and are not required to use ACCRE. Some of this information may be out of date (e.g. setpkgs is no longer supported and has been replaced by Lmod).
- Building, Compiling, and Installing Software
- Setting up your compiling environment
- Searching for symbols
- Determining appropriate compilers
- Determining and adding library paths
- GPU Programming and Computing
- General GPU computing concepts
- Introduction to CUDA
- ACCRE GPU nodes configuration
- GPU job submission
- Advanced SLURM
- Opportunities for using job arrays, their strengths and limitations