feat: add gpu_partition_size to MachineSpec

Google APIs · copybara-github · commit 228264393b90 · 2025-08-12T11:36:44.000-07:00
PiperOrigin-RevId: 794199275
diff --git a/google/cloud/aiplatform/v1beta1/machine_resources.proto b/google/cloud/aiplatform/v1beta1/machine_resources.proto
@@ -55,6 +55,25 @@ message MachineSpec {
   // The number of accelerators to attach to the machine.
   int32 accelerator_count = 3;
 
+  // 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.
+  string gpu_partition_size = 7 [
+    (google.api.field_behavior) = IMMUTABLE,
+    (google.api.field_behavior) = OPTIONAL
+  ];
+
   // Immutable. The topology of the TPUs. Corresponds to the TPU topologies
   // available from GKE. (Example: tpu_topology: "2x2x1").
   string tpu_topology = 4 [(google.api.field_behavior) = IMMUTABLE];
