The contents of your home directory are placed on an AFS name space and is backed up regularly. AFS (Andrew File System) is a distributed networked file system, which was developed for security and scalability. Read more about AFS here.

Most differences between AFS and traditional networked file systems (like NFS) are transparant for the user, but some important differences are:

• AFS has a single systems image. There is the same view of the filestore from each client and server in a network of systems (like HPC2N). This means users can move between machines/workstations and still have the same view of the filestore.
• AFS makes use of Kerberos to authenticate users (this means you need to renew your ticket if you have been logged in for more than 24 hours. Use kinit). AFS also uses access control lists (ACLs) to enable users to restrict access to their own directories.
• AFS uses client caching to reduce network load
• Special AFS Backup system means less system downtime (your files can be found in OldFiles/ after 24 hours).

Note that since ticket-forwarding to batch jobs does not work, the only AFS-access possible from batch jobs are to read files from your Public-directory which is world-wide readable (yes, the entire world) and from the parallel file system. Use the parallel file system for data management in conjunction with batch jobs.

See File systems and Storage for some more information about AFS and the parallel file system.