Compute Canada

How is scratch space handled on the cluster?

Please note: The FAQ pages at the HPCVL website are continuously being revised. Some pages might pertain to an older configuration of the system. Please let us know if you encounter problems or inaccuracies, and we will correct the entries.

Scratch space is supplied in the /scratch area of the file system. This space is intended for transitory data that are generated during a calculation and are usually deleted shortly after the calculation has finished. However, it is worthwhile to consider keeping other intermediate results that are only needed for a short time on scratch space if there is a danger of exceeding disk quota in /home or /u1.

/scratch is subject to a quota of 5 TB per user to avoid sudden overflow on a disk array. If you require more, please contact us.

Note that our scratch space is global, i.e. accessible from all nodes. While this implies somewhat slower access than local scratch, it allows data to be used from different nodes within a program run (e.g. of an MPI program), and it simplifies maintanance.

Scratch space is accessed via the /scratch directory. A directory /scratch/user-ID is automatically created when you receive and HPCVL account. By default, it is only accessible by the owner.

To use the scratch, you will often have to set an application specific environment variable, which can then be given the name /scratch/hpcXXXX to work on all nodes, eg. for the quantum-chemistry codeGaussian, one would set (in a csh):

export GAUSS_SCRDIR=/scratch/hpcXXXX

Note that the above setting is automatically applied when you issue a "use g03" command.