GPUNode

GPUNode is the Schema for the gpunodes API.

Kubernetes Resource Information

Field	Value
API Version	tensor-fusion.ai/v1
Kind	GPUNode
Scope	Cluster

Table of Contents

• API Information
• Spec
• Status

Spec

GPUNodeSpec defines the desired state of GPUNode.

Property	Type	Constraints	Description
cloudVendorParam	string
costPerHour	string
gpuCardIndices	array		if not all GPU cards should be used, specify the GPU card indices, default to empty, onboard all GPU cards to the pool
manageMode	string	Manual AutoSelect Provisioned	Default: AutoSelect

Status

GPUNodeStatus defines the observed state of GPUNode.

Property	Type	Constraints	Description
allocationDetails ↓	array		Allocation details is for node compaction, and calculate used apps
availableTFlops	any	pattern: Regex
availableVRAM	any	pattern: Regex
conditions ↓	array
hypervisorStatus ↓	object
kubernetesNodeName	string		the identifier of the kubernetes node, in nodeSelector mode, GPUNode name is the same as kubernetes node name because of it's owned by the Kubernetes node, while in node provisioning mode owned by the GPUNode, and K8S Node name is uncontrollable
loadedModels	array
managedGPUDeviceIDs	array
managedGPUs	integer<int32>
nodeInfo ↓	object
observedGeneration	integer<int64>
phase	string	Pending Provisioning Migrating Running Succeeded Failed Unknown Destroying	Default: Pending
totalGPUs	integer<int32>
totalTFlops	any	pattern: Regex
totalVRAM	any	pattern: Regex
virtualAvailableTFlops	any	pattern: Regex
virtualAvailableVRAM	any	pattern: Regex
virtualTFlops	any	pattern: Regex
virtualVRAM	any	pattern: Regex

allocationDetails (items)

Allocation details is for node compaction, and calculate used apps

Properties

Property	Type	Constraints	Description
limits ↓	object
namespace	string
podID	string
podName	string
qos	string	low medium high critical
requests ↓	object
workload	string

limits

Properties

Property	Type	Constraints	Description
tflops	any	pattern: Regex	Tera floating point operations per second
vram	any	pattern: Regex	VRAM is short for Video memory, namely GPU RAM

requests

Properties

Property	Type	Constraints	Description
tflops	any	pattern: Regex	Tera floating point operations per second
vram	any	pattern: Regex	VRAM is short for Video memory, namely GPU RAM

conditions (items)

Properties

Property	Type	Constraints	Description
lastTransitionTime	string<date-time>		lastTransitionTime is the last time the condition transitioned from one status to another. This should be when the underlying condition changed. If that is not known, then using the time when the API field changed is acceptable.
message	string	maxLength: 32768	message is a human readable message indicating details about the transition. This may be an empty string.
observedGeneration	integer<int64>	min: 0	observedGeneration represents the .metadata.generation that the condition was set based upon. For instance, if .metadata.generation is currently 12, but the .status.conditions[x].observedGeneration is 9, the condition is out of date with respect to the current state of the instance.
reason	string	minLength: 1 maxLength: 1024 pattern: Regex	reason contains a programmatic identifier indicating the reason for the condition's last transition. Producers of specific condition types may define expected values and meanings for this field, and whether the values are considered a guaranteed API. The value should be a CamelCase string. This field may not be empty.
status	string	True False Unknown	status of the condition, one of True, False, Unknown.
type	string	maxLength: 316 pattern: Regex	type of condition in CamelCase or in foo.example.com/CamelCase.

hypervisorStatus

Properties

Property	Type	Constraints	Description
hypervisorState	string
hypervisorVersion	string
lastHeartbeatTime	string<date-time>

nodeInfo

Properties

Property	Type	Constraints	Description
architecture	string
dataDiskSize	any	pattern: Regex
gpuCount	integer<int32>
gpuDriverVersion	string
gpuModel	string
hostname	string
instanceID	string		only set when node is managed by TensorFusion
ip	string
kernelVersion	string
operatingSystem	string
osImage	string
ramSize	any	pattern: Regex	Additional space for L1/L2 VRAM buffer
region	string