You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Copy file name to clipboardExpand all lines: docs/dyn/aiplatform_v1beta1.batchPredictionJobs.html
+4Lines changed: 4 additions & 0 deletions
Original file line number
Diff line number
Diff line change
@@ -118,6 +118,7 @@ <h3>Method Details</h3>
118
118
"machineSpec": { # Specification of a single machine. # Required. Immutable. The specification of a single machine.
119
119
"acceleratorCount": 42, # The number of accelerators to attach to the machine.
120
120
"acceleratorType": "A String", # Immutable. The type of accelerator(s) that may be attached to the machine as per accelerator_count.
121
+
"gpuPartitionSize": "A String", # Optional. Immutable. The Nvidia GPU partition size. When specified, the requested accelerators will be partitioned into smaller GPU partitions. For example, if the request is for 8 units of NVIDIA A100 GPUs, and gpu_partition_size="1g.10gb", the service will create 8 * 7 = 56 partitioned MIG instances. The partition size must be a value supported by the requested accelerator. Refer to [Nvidia GPU Partitioning](https://cloud.google.com/kubernetes-engine/docs/how-to/gpus-multi#multi-instance_gpu_partitions) for the available partition sizes. If set, the accelerator_count should be set to 1.
121
122
"machineType": "A String", # Immutable. The type of the machine. See the [list of machine types supported for prediction](https://cloud.google.com/vertex-ai/docs/predictions/configure-compute#machine-types) See the [list of machine types supported for custom training](https://cloud.google.com/vertex-ai/docs/training/configure-compute#machine-types). For DeployedModel this field is optional, and the default value is `n1-standard-2`. For BatchPredictionJob or as part of WorkerPoolSpec this field is required.
122
123
"multihostGpuNodeCount": 42, # Optional. Immutable. The number of nodes per replica for multihost GPU deployments.
123
124
"reservationAffinity": { # A ReservationAffinity can be used to configure a Vertex AI resource (e.g., a DeployedModel) to draw its Compute Engine resources from a Shared Reservation, or exclusively from on-demand capacity. # Optional. Immutable. Configuration controlling how this resource pool consumes reservation.
@@ -612,6 +613,7 @@ <h3>Method Details</h3>
612
613
"machineSpec": { # Specification of a single machine. # Required. Immutable. The specification of a single machine.
613
614
"acceleratorCount": 42, # The number of accelerators to attach to the machine.
614
615
"acceleratorType": "A String", # Immutable. The type of accelerator(s) that may be attached to the machine as per accelerator_count.
616
+
"gpuPartitionSize": "A String", # Optional. Immutable. The Nvidia GPU partition size. When specified, the requested accelerators will be partitioned into smaller GPU partitions. For example, if the request is for 8 units of NVIDIA A100 GPUs, and gpu_partition_size="1g.10gb", the service will create 8 * 7 = 56 partitioned MIG instances. The partition size must be a value supported by the requested accelerator. Refer to [Nvidia GPU Partitioning](https://cloud.google.com/kubernetes-engine/docs/how-to/gpus-multi#multi-instance_gpu_partitions) for the available partition sizes. If set, the accelerator_count should be set to 1.
615
617
"machineType": "A String", # Immutable. The type of the machine. See the [list of machine types supported for prediction](https://cloud.google.com/vertex-ai/docs/predictions/configure-compute#machine-types) See the [list of machine types supported for custom training](https://cloud.google.com/vertex-ai/docs/training/configure-compute#machine-types). For DeployedModel this field is optional, and the default value is `n1-standard-2`. For BatchPredictionJob or as part of WorkerPoolSpec this field is required.
616
618
"multihostGpuNodeCount": 42, # Optional. Immutable. The number of nodes per replica for multihost GPU deployments.
617
619
"reservationAffinity": { # A ReservationAffinity can be used to configure a Vertex AI resource (e.g., a DeployedModel) to draw its Compute Engine resources from a Shared Reservation, or exclusively from on-demand capacity. # Optional. Immutable. Configuration controlling how this resource pool consumes reservation.
@@ -1112,6 +1114,7 @@ <h3>Method Details</h3>
1112
1114
"machineSpec": { # Specification of a single machine. # Required. Immutable. The specification of a single machine.
1113
1115
"acceleratorCount": 42, # The number of accelerators to attach to the machine.
1114
1116
"acceleratorType": "A String", # Immutable. The type of accelerator(s) that may be attached to the machine as per accelerator_count.
1117
+
"gpuPartitionSize": "A String", # Optional. Immutable. The Nvidia GPU partition size. When specified, the requested accelerators will be partitioned into smaller GPU partitions. For example, if the request is for 8 units of NVIDIA A100 GPUs, and gpu_partition_size="1g.10gb", the service will create 8 * 7 = 56 partitioned MIG instances. The partition size must be a value supported by the requested accelerator. Refer to [Nvidia GPU Partitioning](https://cloud.google.com/kubernetes-engine/docs/how-to/gpus-multi#multi-instance_gpu_partitions) for the available partition sizes. If set, the accelerator_count should be set to 1.
1115
1118
"machineType": "A String", # Immutable. The type of the machine. See the [list of machine types supported for prediction](https://cloud.google.com/vertex-ai/docs/predictions/configure-compute#machine-types) See the [list of machine types supported for custom training](https://cloud.google.com/vertex-ai/docs/training/configure-compute#machine-types). For DeployedModel this field is optional, and the default value is `n1-standard-2`. For BatchPredictionJob or as part of WorkerPoolSpec this field is required.
1116
1119
"multihostGpuNodeCount": 42, # Optional. Immutable. The number of nodes per replica for multihost GPU deployments.
1117
1120
"reservationAffinity": { # A ReservationAffinity can be used to configure a Vertex AI resource (e.g., a DeployedModel) to draw its Compute Engine resources from a Shared Reservation, or exclusively from on-demand capacity. # Optional. Immutable. Configuration controlling how this resource pool consumes reservation.
@@ -1618,6 +1621,7 @@ <h3>Method Details</h3>
1618
1621
"machineSpec": { # Specification of a single machine. # Required. Immutable. The specification of a single machine.
1619
1622
"acceleratorCount": 42, # The number of accelerators to attach to the machine.
1620
1623
"acceleratorType": "A String", # Immutable. The type of accelerator(s) that may be attached to the machine as per accelerator_count.
1624
+
"gpuPartitionSize": "A String", # Optional. Immutable. The Nvidia GPU partition size. When specified, the requested accelerators will be partitioned into smaller GPU partitions. For example, if the request is for 8 units of NVIDIA A100 GPUs, and gpu_partition_size="1g.10gb", the service will create 8 * 7 = 56 partitioned MIG instances. The partition size must be a value supported by the requested accelerator. Refer to [Nvidia GPU Partitioning](https://cloud.google.com/kubernetes-engine/docs/how-to/gpus-multi#multi-instance_gpu_partitions) for the available partition sizes. If set, the accelerator_count should be set to 1.
1621
1625
"machineType": "A String", # Immutable. The type of the machine. See the [list of machine types supported for prediction](https://cloud.google.com/vertex-ai/docs/predictions/configure-compute#machine-types) See the [list of machine types supported for custom training](https://cloud.google.com/vertex-ai/docs/training/configure-compute#machine-types). For DeployedModel this field is optional, and the default value is `n1-standard-2`. For BatchPredictionJob or as part of WorkerPoolSpec this field is required.
1622
1626
"multihostGpuNodeCount": 42, # Optional. Immutable. The number of nodes per replica for multihost GPU deployments.
1623
1627
"reservationAffinity": { # A ReservationAffinity can be used to configure a Vertex AI resource (e.g., a DeployedModel) to draw its Compute Engine resources from a Shared Reservation, or exclusively from on-demand capacity. # Optional. Immutable. Configuration controlling how this resource pool consumes reservation.
Copy file name to clipboardExpand all lines: docs/dyn/aiplatform_v1beta1.projects.locations.batchPredictionJobs.html
+4Lines changed: 4 additions & 0 deletions
Original file line number
Diff line number
Diff line change
@@ -149,6 +149,7 @@ <h3>Method Details</h3>
149
149
"machineSpec": { # Specification of a single machine. # Required. Immutable. The specification of a single machine.
150
150
"acceleratorCount": 42, # The number of accelerators to attach to the machine.
151
151
"acceleratorType": "A String", # Immutable. The type of accelerator(s) that may be attached to the machine as per accelerator_count.
152
+
"gpuPartitionSize": "A String", # Optional. Immutable. The Nvidia GPU partition size. When specified, the requested accelerators will be partitioned into smaller GPU partitions. For example, if the request is for 8 units of NVIDIA A100 GPUs, and gpu_partition_size="1g.10gb", the service will create 8 * 7 = 56 partitioned MIG instances. The partition size must be a value supported by the requested accelerator. Refer to [Nvidia GPU Partitioning](https://cloud.google.com/kubernetes-engine/docs/how-to/gpus-multi#multi-instance_gpu_partitions) for the available partition sizes. If set, the accelerator_count should be set to 1.
152
153
"machineType": "A String", # Immutable. The type of the machine. See the [list of machine types supported for prediction](https://cloud.google.com/vertex-ai/docs/predictions/configure-compute#machine-types) See the [list of machine types supported for custom training](https://cloud.google.com/vertex-ai/docs/training/configure-compute#machine-types). For DeployedModel this field is optional, and the default value is `n1-standard-2`. For BatchPredictionJob or as part of WorkerPoolSpec this field is required.
153
154
"multihostGpuNodeCount": 42, # Optional. Immutable. The number of nodes per replica for multihost GPU deployments.
154
155
"reservationAffinity": { # A ReservationAffinity can be used to configure a Vertex AI resource (e.g., a DeployedModel) to draw its Compute Engine resources from a Shared Reservation, or exclusively from on-demand capacity. # Optional. Immutable. Configuration controlling how this resource pool consumes reservation.
@@ -642,6 +643,7 @@ <h3>Method Details</h3>
642
643
"machineSpec": { # Specification of a single machine. # Required. Immutable. The specification of a single machine.
643
644
"acceleratorCount": 42, # The number of accelerators to attach to the machine.
644
645
"acceleratorType": "A String", # Immutable. The type of accelerator(s) that may be attached to the machine as per accelerator_count.
646
+
"gpuPartitionSize": "A String", # Optional. Immutable. The Nvidia GPU partition size. When specified, the requested accelerators will be partitioned into smaller GPU partitions. For example, if the request is for 8 units of NVIDIA A100 GPUs, and gpu_partition_size="1g.10gb", the service will create 8 * 7 = 56 partitioned MIG instances. The partition size must be a value supported by the requested accelerator. Refer to [Nvidia GPU Partitioning](https://cloud.google.com/kubernetes-engine/docs/how-to/gpus-multi#multi-instance_gpu_partitions) for the available partition sizes. If set, the accelerator_count should be set to 1.
645
647
"machineType": "A String", # Immutable. The type of the machine. See the [list of machine types supported for prediction](https://cloud.google.com/vertex-ai/docs/predictions/configure-compute#machine-types) See the [list of machine types supported for custom training](https://cloud.google.com/vertex-ai/docs/training/configure-compute#machine-types). For DeployedModel this field is optional, and the default value is `n1-standard-2`. For BatchPredictionJob or as part of WorkerPoolSpec this field is required.
646
648
"multihostGpuNodeCount": 42, # Optional. Immutable. The number of nodes per replica for multihost GPU deployments.
647
649
"reservationAffinity": { # A ReservationAffinity can be used to configure a Vertex AI resource (e.g., a DeployedModel) to draw its Compute Engine resources from a Shared Reservation, or exclusively from on-demand capacity. # Optional. Immutable. Configuration controlling how this resource pool consumes reservation.
@@ -1177,6 +1179,7 @@ <h3>Method Details</h3>
1177
1179
"machineSpec": { # Specification of a single machine. # Required. Immutable. The specification of a single machine.
1178
1180
"acceleratorCount": 42, # The number of accelerators to attach to the machine.
1179
1181
"acceleratorType": "A String", # Immutable. The type of accelerator(s) that may be attached to the machine as per accelerator_count.
1182
+
"gpuPartitionSize": "A String", # Optional. Immutable. The Nvidia GPU partition size. When specified, the requested accelerators will be partitioned into smaller GPU partitions. For example, if the request is for 8 units of NVIDIA A100 GPUs, and gpu_partition_size="1g.10gb", the service will create 8 * 7 = 56 partitioned MIG instances. The partition size must be a value supported by the requested accelerator. Refer to [Nvidia GPU Partitioning](https://cloud.google.com/kubernetes-engine/docs/how-to/gpus-multi#multi-instance_gpu_partitions) for the available partition sizes. If set, the accelerator_count should be set to 1.
1180
1183
"machineType": "A String", # Immutable. The type of the machine. See the [list of machine types supported for prediction](https://cloud.google.com/vertex-ai/docs/predictions/configure-compute#machine-types) See the [list of machine types supported for custom training](https://cloud.google.com/vertex-ai/docs/training/configure-compute#machine-types). For DeployedModel this field is optional, and the default value is `n1-standard-2`. For BatchPredictionJob or as part of WorkerPoolSpec this field is required.
1181
1184
"multihostGpuNodeCount": 42, # Optional. Immutable. The number of nodes per replica for multihost GPU deployments.
1182
1185
"reservationAffinity": { # A ReservationAffinity can be used to configure a Vertex AI resource (e.g., a DeployedModel) to draw its Compute Engine resources from a Shared Reservation, or exclusively from on-demand capacity. # Optional. Immutable. Configuration controlling how this resource pool consumes reservation.
@@ -1683,6 +1686,7 @@ <h3>Method Details</h3>
1683
1686
"machineSpec": { # Specification of a single machine. # Required. Immutable. The specification of a single machine.
1684
1687
"acceleratorCount": 42, # The number of accelerators to attach to the machine.
1685
1688
"acceleratorType": "A String", # Immutable. The type of accelerator(s) that may be attached to the machine as per accelerator_count.
1689
+
"gpuPartitionSize": "A String", # Optional. Immutable. The Nvidia GPU partition size. When specified, the requested accelerators will be partitioned into smaller GPU partitions. For example, if the request is for 8 units of NVIDIA A100 GPUs, and gpu_partition_size="1g.10gb", the service will create 8 * 7 = 56 partitioned MIG instances. The partition size must be a value supported by the requested accelerator. Refer to [Nvidia GPU Partitioning](https://cloud.google.com/kubernetes-engine/docs/how-to/gpus-multi#multi-instance_gpu_partitions) for the available partition sizes. If set, the accelerator_count should be set to 1.
1686
1690
"machineType": "A String", # Immutable. The type of the machine. See the [list of machine types supported for prediction](https://cloud.google.com/vertex-ai/docs/predictions/configure-compute#machine-types) See the [list of machine types supported for custom training](https://cloud.google.com/vertex-ai/docs/training/configure-compute#machine-types). For DeployedModel this field is optional, and the default value is `n1-standard-2`. For BatchPredictionJob or as part of WorkerPoolSpec this field is required.
1687
1691
"multihostGpuNodeCount": 42, # Optional. Immutable. The number of nodes per replica for multihost GPU deployments.
1688
1692
"reservationAffinity": { # A ReservationAffinity can be used to configure a Vertex AI resource (e.g., a DeployedModel) to draw its Compute Engine resources from a Shared Reservation, or exclusively from on-demand capacity. # Optional. Immutable. Configuration controlling how this resource pool consumes reservation.
0 commit comments