Class RuntimeUpdatableParams (0.13.0)

RuntimeUpdatableParams(mapping=None, *, ignore_unknown_fields=False, **kwargs)

Additional job parameters that can only be updated during runtime using the projects.jobs.update method. These fields have no effect when specified during job creation.

.. _oneof: https://proto-plus-python.readthedocs.io/en/stable/fields.html#oneofs-mutually-exclusive-fields

Attributes

Name Description
max_num_workers int
The maximum number of workers to cap autoscaling at. This field is currently only supported for Streaming Engine jobs. This field is a member of oneof_ _max_num_workers.
min_num_workers int
The minimum number of workers to scale down to. This field is currently only supported for Streaming Engine jobs. This field is a member of oneof_ _min_num_workers.
worker_utilization_hint float
Target worker utilization, compared against the aggregate utilization of the worker pool by autoscaler, to determine upscaling and downscaling when absent other constraints such as backlog. For more information, see `Update an existing pipeline
acceptable_backlog_duration google.protobuf.duration_pb2.Duration
Optional. Deprecated: Use autoscaling_tier instead. The backlog threshold duration in seconds for autoscaling. Value must be non-negative. This field is a member of oneof_ _acceptable_backlog_duration.
autoscaling_tier str
Optional. The backlog threshold tier for autoscaling. Value must be one of "low-latency", "medium-latency", or "high-latency". This field is a member of oneof_ _autoscaling_tier.