srun Option

Description

--uid=user

Root attempts to submit or run a job as

 

normal user.

-t

Establish a time limit to terminate the job

--time=minutes

after specified number of minutes.

 

--gid=group

Root attempts to submit or run a job as

 

group.

-A

Allocate resource and spawn a shell.

--allocate

 

--no-shell

Immediately exit after allocating resources.

-I

Allocate immediately or fail.

--immediate

 

-s

Share nodes with other running jobs.

--share

SHARED=FORCE shares all nodes in

 

partition.

 

SHARED=YES shares nodes if and only if

 

–shareis specified.

 

SHARED=NO means do not share the node.

-O

Overcommit resources.

--overcommit

 

-b

Submit in “batch mode”.

--batch

 

-r

Run a job step relative to node n of the

--relative=n

current allocation. It is about placing tasks

within allocation.

 

-D

Specify the working directory of the job.

--chdir=path

 

-k

Do not automatically terminate a job if one

--no-kill

of the nodes it has been allocated fails.

 

-T

The srun command uses nthreads to

--threads=nthreads

initiate and control the parallel job.

 

-l

Prepend task number to lines of

--label

stdout/stderr.

 

-u

Do not line buffer stdout from remote tasks.

--unbuffered

 

-m

Distribution method for remote processes.

--distribution=blockcyl

 

--mpi=mpi_type

Identify the type of MPI to be used.

--jobid=id

Initiate a job step under an already allocated

 

job id.

-v

Verbose operation.

-d

A debug level for slurmd.

--slurm-debug=level

 

LSF-HPC Equivalent

You cannot use this option. LSF-HPC uses it to create allocation.

bsub -W runlimit

You cannot use this option. LSF-HPC uses this option to create allocation.

You cannot use this option. LSF-HPC uses this option to create allocation.

You cannot use this option. LSF-HPC uses this option to create allocation.

You cannot use this option. LSF-HPC uses this option to create allocation.

You cannot use this option. LSF-HPC uses this option to create allocation.

Use when launching parallel tasks.

Meaningless under LSF-HPC.

Use as an argument to srun when launching parallel tasks.

Job starts in job submission directory by default.

When creating allocation, LSF-HPC does not use the –koption. SLURM always terminates a job if one of allocated nodes fails

Use as an argument to srun when launching parallel tasks.

Use as an argument to srun when launching parallel tasks.

Use as an argument to srun when launching parallel tasks.

Use as an argument to srun when launching parallel tasks.

Use as an argument to srun when launching parallel tasks.

Do not use. SLURM_JOBID is set in job environment to associate job with SLURM allocation.

Use as an argument to srun when launching parallel tasks.

Use as an argument to srun when launching parallel tasks.

88 Using LSF

Page 88
Image 88
HP XC System 3.x Software manual Mpi=mpitype