Access your files
The cluster has a parallel file system which provides fast and reliable access to your data. You should have at least 3 file sets:
- User home directory
- This is where you store your persistent data
- Located at /pfss/home/$USER
- The default quota is 10GB
- User scratch directory
- This is where you use for I/O during your job
- Located at /pfss/scratch01/$USER
- The default quota is 100GB
- Inactive files may be purged every 30 days
- Group scratch directory
- This is where you share your file with your group mate
- Located at/pfss/scratch02/$GROUP
- The default quota is 1TB
- This is where you share your file with your group mate
There are three different ways to download or upload your files.
Web portal
We built a file browser in the portal. You may open it up by clicking files in the top menu bar.
You may switch between different file sets by clicking the hard disk icon. The quota and consumption are displayed on top of that. In the top right corner, there is a button for you to upload files. You may upload multiple files at once under a limit of 10GB.
Click on the item name to enter a directory or download a file. If it is a .sbatch file, the job launcher will be fired for enqueuing a new job. There is a dropdown menu on the right-hand side to copy the path, rename, or delete an item.