Editing HPC Terminology (section)

=== Local Scratch Partition ===

When you are frequently accessing large files, you should first move them to the <code>/scratch</code> directory of the local machines your job is running on. Read/Write is much faster locally than it is over the network.

The <code>/scratch</code> dir of each node is exported via NFS to other nodes on the cluster. The local disk for the node <code>&lt;node_name&gt;</code> can be accessed through at <code>/compute/&lt;node_name&gt;</code> on other nodes. This will allow for faster access and reduce pressure on the NAS. ''(highly recommended for large or frequently accessed files)''.

'''Notes''':

* If you <code>ls /compute/</code>, it might seem empty. However if you add the compute node's hostname the path autofs will mount the remote directory and you will be able to access it's <code>scratch</code> partition. For example:

    [dvosler@hpc-1-27 ~]# ls -la /compute/
    drwxr-xr-x   2 root root    0 Jun 23 15:02 .
    dr-xr-xr-x. 21 root root 4096 Jun 23 15:02 ..

    [dvosler@hpc-1-27 ~]# ls -la /compute/hpc-0-19
    drwxrwxrwt 4 root     root      4096 May 29 10:25 .
    drwxr-xr-x 3 root     root         0 Jun 28 16:17 ..
    drwxrwxr-x 3 dvosler  dvosler   4096 May 11 18:41 things
    -rw-rw-r-- 3 dvosler  dvosler    420 May 11 18:41 and
    -rw-rw-r-- 3 lemmy    hpcusers     1 May 11 18:41 stuff

* Compute nodes and scratch should only hold temporary files, in the sense that you should clean up after yourself. If scratch is 85% full, data older than 28 days will be deleted automatically.