The best practice to avoid problems with overload and slow response of NFS servers is: read all input files from the NFS server only once when your job starts, use /scratch at a local disk of an assigned worker node as your working directory, write all output files back to the NFS server when your job completes.
A local working directory /scratch is available at a local disk of each worker node, and is intended for temporary files. Upon a job start, the batch system creates a unique subdirectory /scratch/condor/dir_ID/, which is accessible via the $_CONDOR_SCRATCH_DIR variable. When the job finishes, the subdirectory is automatically deleted. This way, the local working directory /scratch is continuously cleaned.
Add a requirement for /scratch to a job_submission_file.
request_disk = 20G| Worker Node Name | Total /scratch | /scratch per one HTcore |
|---|---|---|
| milan01-29 | 3.3 TB | 25.7 GB |
| minis01-12 | 722 GB | 22.5 GB |
| mahagon01-24 | 1.6 TB | 25 GB |
| mikan01-51 | 941 GB | 19.6 GB |
English 
Čeština