HP XC System Software User's Guide Version 3.2

Table 10-3 LSF-HPC Equivalents of SLURM srun Options (continued)
LSF-HPC EquivalentDescriptionsrun Option
You can use when launching parallel tasks.
Specify how stdin is to be redirected.-i
--input=none,tasked
bsub -e error_fileSpecify how stderr is to be
redirected.
-e
--error=file_name
You can use when launching parallel tasks.
Specify how stderr is to be
redirected.
-e
--error=none,tasked
You cannot use this option. When creating
allocation. SLURM sets LSF-HPC job id
automatically.
Specify a name for the job.
-J
--job-name=job_name
You cannot use this option. LSF-HPC uses it to
create allocation.
Root attempts to submit or run a job
as normal user.
--uid=user
bsub -W runlimit
Establish a time limit to terminate the
job after specified number of minutes.
-t
--time=minutes
You cannot use this option. LSF-HPC uses this
option to create allocation.
Root attempts to submit or run a job
as group.
--gid=group
You cannot use this option. LSF-HPC uses this
option to create allocation.
Allocate resource and spawn a shell.
-A
--allocate
You cannot use this option. LSF-HPC uses this
option to create allocation.
Immediately exit after allocating
resources.
--no-shell
You cannot use this option. LSF-HPC uses this
option to create allocation.
Allocate immediately or fail.
-I
--immediate
You cannot use this option. LSF-HPC uses this
option to create allocation.
Share nodes with other running jobs.
SHARED=FORCE shares all nodes in
partition.
SHARED=YES shares nodes if and only
if –share is specified.
SHARED=NO means do not share the
node.
-s
--share
Use when launching parallel tasks.Overcommit resources.
-O
--overcommit
Meaningless under LSF-HPC integrated with
SLURM.
Submit in “batch mode”.
-b
--batch
Use as an argument to srun when launching
parallel tasks.
Run a job step relative to node n of the
current allocation. It is about placing
tasks within allocation.
-r
--relative=n
Job starts in job submission directory by
default.
Specify the working directory of the
job.
-D
--chdir=path
When creating allocation, LSF-HPC does not
use the –k option. SLURM always terminates
a job if one of allocated nodes fails
Do not automatically terminate a job
if one of the nodes it has been allocated
fails.
-k
--no-kill
Use as an argument to srun when launching
parallel tasks.
The srun command uses nthreads
to initiate and control the parallel job.
-T
--threads=nthreads
10.14 LSF-HPC Equivalents of SLURM srun Options 115