pulumi-google-native-kotlin/com.pulumi.googlenative.dataproc.v1.kotlin.inputs/GkeNodePoolAcceleratorConfigArgs

GkeNodePoolAcceleratorConfigArgs

data class GkeNodePoolAcceleratorConfigArgs(val acceleratorCount: Output<String>? = null, val acceleratorType: Output<String>? = null, val gpuPartitionSize: Output<String>? = null) : ConvertibleToJava<GkeNodePoolAcceleratorConfigArgs>

A GkeNodeConfigAcceleratorConfig represents a Hardware Accelerator request for a node pool.

Constructors

Link copied to clipboard
fun GkeNodePoolAcceleratorConfigArgs(acceleratorCount: Output<String>? = null, acceleratorType: Output<String>? = null, gpuPartitionSize: Output<String>? = null)

Functions

Link copied to clipboard
open override fun toJava(): GkeNodePoolAcceleratorConfigArgs

Properties

Link copied to clipboard
val acceleratorCount: Output<String>? = null

The number of accelerator cards exposed to an instance.

Link copied to clipboard
val acceleratorType: Output<String>? = null

The accelerator type resource namename (see GPUs on Compute Engine).

Link copied to clipboard
val gpuPartitionSize: Output<String>? = null

Size of partitions to create on the GPU. Valid values are described in the NVIDIA mig user guide (https://docs.nvidia.com/datacenter/tesla/mig-user-guide/#partitioning).