Reference documentation and code samples for the Cloud Dataproc V1 API class Google::Cloud::Dataproc::V1::GkeNodePoolConfig::GkeNodePoolAcceleratorConfig.
A GkeNodeConfigAcceleratorConfig represents a Hardware Accelerator request for a node pool.
Inherits
- Object
Extended By
- Google::Protobuf::MessageExts::ClassMethods
Includes
- Google::Protobuf::MessageExts
Methods
#accelerator_count
def accelerator_count() -> ::Integer
Returns
- (::Integer) — The number of accelerator cards exposed to an instance.
#accelerator_count=
def accelerator_count=(value) -> ::Integer
Parameter
- value (::Integer) — The number of accelerator cards exposed to an instance.
Returns
- (::Integer) — The number of accelerator cards exposed to an instance.
#accelerator_type
def accelerator_type() -> ::String
Returns
- (::String) — The accelerator type resource namename (see GPUs on Compute Engine).
#accelerator_type=
def accelerator_type=(value) -> ::String
Parameter
- value (::String) — The accelerator type resource namename (see GPUs on Compute Engine).
Returns
- (::String) — The accelerator type resource namename (see GPUs on Compute Engine).
#gpu_partition_size
def gpu_partition_size() -> ::String
Returns
- (::String) — Size of partitions to create on the GPU. Valid values are described in the NVIDIA mig user guide.
#gpu_partition_size=
def gpu_partition_size=(value) -> ::String
Parameter
- value (::String) — Size of partitions to create on the GPU. Valid values are described in the NVIDIA mig user guide.
Returns
- (::String) — Size of partitions to create on the GPU. Valid values are described in the NVIDIA mig user guide.