@@ -195,6 +195,18 @@ message BatchDedicatedResources {
195195 // Immutable. The maximum number of machine replicas the batch operation may
196196 // be scaled to. The default value is 10.
197197 int32 max_replica_count = 3 [(google.api.field_behavior ) = IMMUTABLE ];
198+
199+ // Optional. Immutable. If set, use DWS resource to schedule the deployment
200+ // workload. reference:
201+ // (https://cloud.google.com/blog/products/compute/introducing-dynamic-workload-scheduler)
202+ FlexStart flex_start = 4 [
203+ (google.api.field_behavior ) = IMMUTABLE ,
204+ (google.api.field_behavior ) = OPTIONAL
205+ ];
206+
207+ // Optional. If true, schedule the deployment workload on [spot
208+ // VMs](https://cloud.google.com/kubernetes-engine/docs/concepts/spot-vms).
209+ bool spot = 5 [(google.api.field_behavior ) = OPTIONAL ];
198210}
199211
200212// Statistics information about resource consumption.
@@ -256,6 +268,7 @@ message AutoscalingMetricSpec {
256268 // * For Online Prediction:
257269 // * `aiplatform.googleapis.com/prediction/online/accelerator/duty_cycle`
258270 // * `aiplatform.googleapis.com/prediction/online/cpu/utilization`
271+ // * `aiplatform.googleapis.com/prediction/online/request_count`
259272 string metric_name = 1 [(google.api.field_behavior ) = REQUIRED ];
260273
261274 // The target resource utilization in percentage (1% - 100%) for the given
0 commit comments