...
A general recommendation for network filesystems is to keep the number of metadata operations for open and closing files, as well as checks for file existence or changes as low as possible. These operations often become a bottleneck for the IO of your job and on large clusters, as the ones operated by HLRN, can easily overload the file servers.
Separate
...
GPU-scratch for grete Partitions on Emmy
On Emmy, glogin9.hlrn.de and the Grete nodes (ggpu[101-134,146-147,201-202]) have their own separate /scratch
that is different from the one on the other frontend and compute nodes. On these nodes, you can still access the
/scratch
of the other frontend and compute nodes at /scratch-emmy
.
...