The backend job scheduler is Gridengine, which is different from the PBS used by NCI, but both work similarly
Resource Requests
The table below summarises the major resources attributes which are commonly used in most of the jobs. There are also other attributes which are very helping for fine tuning how a job should be scheduled, whose details are provided in the sub sections.
Resources | Attribute | Description | Default Value |
|---|---|---|---|
Parallel Environments ( | smp | Allocate X number of CPUs on the SAME compute node | |
mpi | Allocate X number of CPUs from multiple compute node, this is mainly used by a job that implements under Open MPI framework. | ||
Resource request list ( | mem | The amount of memory will be used by a job | |
jobfs | The amount of disk space will be used by a job | ||
walltime | The run time limit for a job | ||
ngpus | The number of GPGPUs | ||
GPGPU
TODO
Rack Awareness
TODO
Submit a Batch Job
A batch job can be submitted by using command qsub, in the following pattern:
qsub -N JOB_NAME -pe smp NUMBER_OF_CPU -l ATTR1=VAL1,ATTR2=VAL2 SCRIPT
Examples
# a very big sleep job that needs 16 x CPUs, 2 x GPGPUs, 64GB memory, 10G disk space qsub -b y -N generic_gpgpu -pe smp 16 -l ngpus=2,mem=65G,jobfs=10G sleep 1m # a smaller sleep job that requires the specific A2 GPGPU... qsub -b y -N t1000_gpgpu -pe smp 8 -l ngpus=2,gpgpu_model=A2,mem=16G,jobfs=10G sleep 1m # a big job runs on multiple H100 nodes inside the same physical rack/cabinet F (rack awareness) qsub -b y -N h100_gpgpu -pe mpi 256 -l ngpus=2,gpgpu_model=H100,rack=f,mem=128G,jobfs=100G sleep 1m
Submission Script
#!/bin/bash # # It prints the actual path of the job scratch directory. #$ -pe smp 8 #$ -j y #$ -e logs/$JOB_ID_$JOB_NAME.out #$ -o logs/$JOB_ID_$JOB_NAME.out #$ -cwd #$ -N dd_smp #$ -l mem=1G,jobfs=110G,tmpfree=150G,walltime=00:30:00 # echo "$HOST $tmp_requested $TMPDIR" # about 107GB dd if=/dev/zero of=$TMPDIR/dd.test bs=512M count=200