package temporal.api.compute.v1

Get desktop application:
View/edit binary Protocol Buffers messages

ComputeConfig stores configuration that helps a worker control plane controller understand *when* and *how* to respond to worker lifecycle events.

Used in: deployment.v1.WorkerDeploymentVersionInfo, workflowservice.v1.CreateWorkerDeploymentVersionRequest

• map<string, ComputeConfigScalingGroup> scaling_groups = 1

  Each scaling group describes a compute config for a specific subset of the worker deployment version: covering a specific set of task types and/or regions. Having different configurations for different task types, allows independent tuning of activity and workflow task processing (for example). The key of the map is the ID of the scaling group used to reference it in subsequent update calls.

Used in: ComputeConfig, ComputeConfigScalingGroupUpdate

• repeated enums.v1.TaskQueueType task_queue_types = 1

  Optional. The set of task queue types this scaling group serves. If not provided, this scaling group serves all not otherwise defined task types.

• optional ComputeProvider provider = 3

  Stores instructions for a worker control plane controller how to respond to worker lifeycle events.

• optional ComputeScaler scaler = 4

  Informs a worker lifecycle controller *when* and *how often* to perform certain worker lifecycle actions like starting a serverless worker.

Used in: ComputeConfigSummary

• repeated enums.v1.TaskQueueType task_queue_types = 1

• string provider_type = 2

Used in: workflowservice.v1.UpdateWorkerDeploymentVersionComputeConfigRequest, workflowservice.v1.ValidateWorkerDeploymentVersionComputeConfigRequest

• optional ComputeConfigScalingGroup scaling_group = 1

• optional google.protobuf.FieldMask update_mask = 2

  Controls which fields from `scaling_group` will be applied. Semantics: - Mask is ignored for new scaling groups (only applicable when scaling group already exists). - Empty mask for an existing scaling group is no-op: no change. - Non-empty mask for an existing scaling group will update/unset only to the fields mentioned in the mask. - Accepted paths: "task_queue_types", "provider", "provider.type", "provider.details", "provider.nexus_endpoint", "scaler", "scaler.type", "scaler.details"

A subset of information in ComputeConfig optimized for list views.

Used in: deployment.v1.WorkerDeploymentInfo.WorkerDeploymentVersionSummary

• map<string, ComputeConfigScalingGroupSummary> scaling_groups = 1

ComputeProvider stores information used by a worker control plane controller to respond to worker lifecycle events. For example, when a Task is received on a TaskQueue that has no active pollers, a serverless worker lifecycle controller might need to invoke an AWS Lambda Function that itself ends up calling the SDK's worker.New() function.

Used in: ComputeConfigScalingGroup

• string type = 1

  Type of the compute provider. This string is implementation-specific and can be used by implementations to understand how to interpret the contents of the provider_details field.

• optional common.v1.Payload details = 2

  Contains provider-specific instructions and configuration. For server-implemented providers, use the SDK's default content converter to ensure the server can understand it. For remote-implemented providers, you might use your own content converters according to what the remote endpoints understand.

• string nexus_endpoint = 10

  Optional. If the compute provider is a Nexus service, this should point there.

ComputeScaler instructs the Temporal Service when to scale up or down the number of Workers that comprise a WorkerDeployment.

Used in: ComputeConfigScalingGroup

• string type = 1

  Type of the compute scaler. this string is implementation-specific and can be used by implementations to understand how to interpret the contents of the scaler_details field.

• optional common.v1.Payload details = 2

  Contains scaler-specific instructions and configuration. For server-implemented scalers, use the SDK's default data converter to ensure the server can understand it. For remote-implemented scalers, you might use your own data converters according to what the remote endpoints understand.