nemo_microservices.types.jobs.volcano_job_execution_profile_config#
Module Contents#
Classes#
API#
- class nemo_microservices.types.jobs.volcano_job_execution_profile_config.VolcanoJobExecutionProfileConfig(/, **data: Any)#
Bases:
nemo_microservices._models.BaseModel- active_deadline_seconds: int | None#
None
- affinity: Dict[str, object] | None#
None
- image_pull_secrets: List[nemo_microservices.types.jobs.kubernetes_job_image_pull_secret.KubernetesJobImagePullSecret] | None#
None
- job_metadata: nemo_microservices.types.jobs.kubernetes_object_metadata.KubernetesObjectMetadata | None#
None
- logging: nemo_microservices.types.jobs.kubernetes_logging_sidecar.KubernetesLoggingSidecar | None#
None
Configuration for Kubernetes Logging Sidecar
- max_retry: int | None#
None
maxRetry indicates the maximum number of retries allowed by the job
- namespace: str | None#
None
Kubernetes namespace to submit the job to.
If not set, it will be determined from the environment.
- node_selector: Dict[str, str] | None#
None
- num_gpus: int | None#
None
- plugins: Dict[str, object] | None#
None
plugins indicates the plugins used by Volcano when the job is scheduled.
We always add the pytorch plugin if more than one node.
- pod_metadata: nemo_microservices.types.jobs.kubernetes_object_metadata.KubernetesObjectMetadata | None#
None
- queue: str | None#
None
The Volcano queue to submit the job to.
- resources: nemo_microservices.types.jobs.compute_resources.ComputeResources | None#
None
Resource requirements matching k8s ResourceRequirements format.
- scheduler_name: str | None#
None
The scheduler name to use for the Volcano job.
- storage: nemo_microservices.types.jobs.kubernetes_job_storage_config.KubernetesJobStorageConfig#
None
Configuration for persistent storage in Kubernetes jobs.
- tolerations: List[Dict[str, object]] | None#
None
- ttl_seconds_after_finished: int | None#
None