Page tree
Skip to end of metadata
Go to start of metadata

You are viewing an old version of this page. View the current version.

Compare with Current View Page History

« Previous Version 4 Current »

Backed up. 10 GiB fixed quota per user.
Not backed up, temporary files, auto purge policy applied.
Not backed up, long-term large data files.
Read only, centrally installed software applications and their module files.
Not backed up, local to the node, I/O intensive data.
Backed up, archiving large data files.
   ○ man mdss
Read manual of all `mdss` commands
   ○ mdss dmls -l
List files with status: online/in disk cache(REG), on tape(OFF), or both (DUL) 
   ○ mdss put/get
Put or retrieve files from mdss
Display compute grant and usage
quota -s
Display storage grant and usage on /home
Display storage grant and usage on /scratch and /g/data by project
nci-files-report -f <fs> -g <grp> 
Report data usage on filesystem <fs> owned by the project <grp> by (project folder, user) 
Module command
module avail
List all available modules
module load <app>/<ver>
Load the module for application <app> version <ver>
module unload <app>
Remove the module for application <app> from the current shell
module show <app>/<ver>
Show what the module <app>/<version> does
module list
List all modules loaded in the current shell
PBSPro commandRun commands in this section with the frequency < 0.2 Hz
Submit job defined in the submission script
qdel <jobid>
Delete the job with jobID <jobid>
qstat -swx <jobid>
Display the job status in the queue with comment
qstat -fx <jobid>
Display full job status information 
qps <jobid>

Take a snapshot of the process status of all current processes in the running job 

qcat [-s/-o/-e] <jobid>
Display [submission script/STDOUT/STDERR] of the running job 
qls <jobid>

List contents in the folder $PBS_JOBFS 

qcp <jobid> <dst>

Copy files and directories from the folder $PBS_JOBFS to the destination folder <dst>

PBSPro Directives
#PBS -P <prj>
Project for job debiting, /scratch project folder access and data ownership   
#PBS -q <queue>
Submit the job to the queue <queue>
#PBS -l ncpus=<xx>
Request <xx> CPU cores
#PBS -l ncpus=<scratch/prj1+gdata/prj2+massdata/prj3>
Storage needed to be available inside the job. massdata is only available in copyq jobs.
#PBS -l ngpus=<yy>
Number of GPUs, ncpus has to be 12 x ngpus and the job has to be submitted to `gpuvolta`.
#PBS -l walltime=<hh:mm:ss>
Max walltime the job would run
#PBS -l mem=<10GB>
Memory allocation
#PBS -l jobfs=<40GB>
Disk allocation on compute/copyq node(s)
#PBS -l software=<app1,app2>
Licenses required
#PBS -l wd
Start the job from the directory in which it was submitted
#PBS -W depend=beforeok:<jobid1,jobid2>
Set dependencies between this and other jobs.
#PBS -a <YYMMhhmm>
Time after which the job is eligible for execution
#PBS -M <,>
List of receivers to whom email about the job is sent
#PBS -m <aben>
Email events. `a` for abortioin, `b` for begin, `e` for end, `n` for none
Useful URLs
User & Project
Service Desk
License Live Status

  • No labels