Resources parameters

Note

One can load, modify, and export the input file by using our effective web-based tool DP-GUI. All parameters below can be set in DP-GUI. By clicking “SAVE JSON”, one can download the input file for.

resources:

type: dict

argument path: resources

number_node:

type: int, optional, default: 1

argument path: resources/number_node

The number of node need for each job

cpu_per_node:

type: int, optional, default: 1

argument path: resources/cpu_per_node

cpu numbers of each node assigned to each job.

gpu_per_node:

type: int, optional, default: 0

argument path: resources/gpu_per_node

gpu numbers of each node assigned to each job.

queue_name:

type: str, optional, default: (empty string)

argument path: resources/queue_name

The queue name of batch job scheduler system.

group_size:

type: int

argument path: resources/group_size

The number of tasks in a job. 0 means infinity.

custom_flags:

type: list, optional

argument path: resources/custom_flags

The extra lines pass to job submitting script header

strategy:

type: dict, optional

argument path: resources/strategy

strategies we use to generation job submitting scripts.

if_cuda_multi_devices:

type: bool, optional, default: False

argument path: resources/strategy/if_cuda_multi_devices

If there are multiple nvidia GPUS on the node, and we want to assign the tasks to different GPUS.If true, dpdispatcher will manually export environment variable CUDA_VISIBLE_DEVICES to different task.Usually, this option will be used with Task.task_need_resources variable simultaneously.

ratio_unfinished:

type: float, optional, default: 0.0

argument path: resources/strategy/ratio_unfinished

The ratio of jobs that can be unfinished.

para_deg:

type: int, optional, default: 1

argument path: resources/para_deg

Decide how many tasks will be run in parallel.

source_list:

type: list, optional, default: []

argument path: resources/source_list

The env file to be sourced before the command execution.

module_purge:

type: bool, optional, default: False

argument path: resources/module_purge

Remove all modules on HPC system before module load (module_list)

module_unload_list:

type: list, optional, default: []

argument path: resources/module_unload_list

The modules to be unloaded on HPC system before submitting jobs

module_list:

type: list, optional, default: []

argument path: resources/module_list

The modules to be loaded on HPC system before submitting jobs

envs:

type: dict, optional, default: {}

argument path: resources/envs

The environment variables to be exported on before submitting jobs

prepend_script:

type: list, optional, default: []

argument path: resources/prepend_script

Optional script run before jobs submitted.

append_script:

type: list, optional, default: []

argument path: resources/append_script

Optional script run after jobs submitted.

wait_time:

type: int | float, optional, default: 0

argument path: resources/wait_time

The waitting time in second after a single task submitted

Depending on the value of batch_type, different sub args are accepted.

batch_type:

type: str (flag key)

argument path: resources/batch_type

possible choices: Torque, Shell, Slurm, PBS, Bohrium, SlurmJobArray, LSF, DistributedShell

The batch job system type loaded from machine/batch_type.

When batch_type is set to Torque (or its alias torque):

kwargs:

type: dict, optional

argument path: resources[Torque]/kwargs

This field is empty for this batch.

When batch_type is set to Shell (or its alias shell):

kwargs:

type: dict, optional

argument path: resources[Shell]/kwargs

This field is empty for this batch.

When batch_type is set to Slurm (or its alias slurm):

kwargs:

type: dict, optional

argument path: resources[Slurm]/kwargs

Extra arguments.

custom_gpu_line:

type: NoneType | str, optional, default: None

argument path: resources[Slurm]/kwargs/custom_gpu_line

Custom GPU configuration, starting with #SBATCH

When batch_type is set to PBS (or its alias pbs):

kwargs:

type: dict, optional

argument path: resources[PBS]/kwargs

This field is empty for this batch.

When batch_type is set to Bohrium (or its aliases bohrium, Lebesgue, lebesgue, DpCloudServer, dpcloudserver):

kwargs:

type: dict, optional

argument path: resources[Bohrium]/kwargs

This field is empty for this batch.

When batch_type is set to SlurmJobArray (or its alias slurmjobarray):

kwargs:

type: dict, optional

argument path: resources[SlurmJobArray]/kwargs

Extra arguments.

custom_gpu_line:

type: NoneType | str, optional, default: None

argument path: resources[SlurmJobArray]/kwargs/custom_gpu_line

Custom GPU configuration, starting with #SBATCH

When batch_type is set to LSF (or its alias lsf):

kwargs:

type: dict

argument path: resources[LSF]/kwargs

Extra arguments.

gpu_usage:

type: bool, optional, default: False

argument path: resources[LSF]/kwargs/gpu_usage

Choosing if GPU is used in the calculation step.

gpu_new_syntax:

type: bool, optional, default: False

argument path: resources[LSF]/kwargs/gpu_new_syntax

For LFS >= 10.1.0.3, new option -gpu for #BSUB could be used. If False, and old syntax would be used.

gpu_exclusive:

type: bool, optional, default: True

argument path: resources[LSF]/kwargs/gpu_exclusive

Only take effect when new syntax enabled. Control whether submit tasks in exclusive way for GPU.

custom_gpu_line:

type: NoneType | str, optional, default: None

argument path: resources[LSF]/kwargs/custom_gpu_line

Custom GPU configuration, starting with #BSUB

When batch_type is set to DistributedShell (or its alias distributedshell):

kwargs:

type: dict, optional

argument path: resources[DistributedShell]/kwargs

This field is empty for this batch.