default to compactFirst for cost saving and energy saving.

in some cases, use lowLoadFirst for balance and fairness.

These are the valid phases of a GpuConnection.

These are the valid phases of a GpuConnection.

These are the valid phases of a GpuConnection.

AddToScheme adds the types in this group-version to the given scheme.

GroupVersion is group version used to register these objects.

SchemeBuilder is used to add go types to the GroupVersionKind scheme.

A typical autoLimits algorithm could be checking every 5m, look back 1 day data, select 99% of actual usage as preferredLimits, calculate finalPreferredLimits, which is preferredLimits*(1+extraBufferRatio) if they are equal with each other within a range (eg.

To handle burst traffic, scale up in short time (this feature requires GPU context migration & replication, not available yet).

Customize system components for seamless onboarding.

ComputingVendorConfig defines the Cloud vendor connection such as AWS, GCP, Azure etc.

DataPipelinesConfig defines the aggregation jobs that can make statistics on the data and then report to cloud if configured.

Enroll to TensorFusion cloud with a enrollment key.

+kubebuilder:object:root=true +kubebuilder:subresource:status +kubebuilder:resource:scope=Cluster GPU is the Schema for the gpus API.

GPUFilter is to select eligible GPUs for scheduling.

GPUList contains a list of GPU.

GPUNode is the Schema for the gpunodes API.

GPUNodeClass is the Schema for the gpunodeclasses API.

GPUNodeClassList contains a list of GPUNodeClass.

GPUNodeClassSpec defines the desired state of GPUNodeClass.

GPUNodeClassStatus defines the observed state of GPUNodeClass.

GPUNodeList contains a list of GPUNode.

GPUNodeSpec defines the desired state of GPUNode.

GPUNodeStatus defines the observed state of GPUNode.

GPUPool is the Schema for the gpupools API.

GPUPool defines how to create a GPU pool, could be URL or inline.

GPUPoolList contains a list of GPUPool.

GPUPoolSpec defines the desired state of GPUPool.

GPUPoolStatus defines the observed state of GPUPool.

GPUStatus defines the observed state of GPU.

NodeProvisioner or NodeSelector, they are exclusive.

Define different QoS and their price.

RemoteWriteConfig represents the configuration for remote write.

SchedulingConfigTemplate is the Schema for the schedulingconfigtemplates API.

SchedulingConfigTemplateList contains a list of SchedulingConfigTemplate.

Place the workload to right nodes and scale smart.

SchedulingConfigTemplateStatus defines the observed state of SchedulingConfigTemplate.

StorageVendorConfig defines Postgres database with extensions for timeseries storage and other resource aggregation results, system events and diagnostics reports etc.

TensorFusionCluster is the Schema for the tensorfusionclusters API.

TensorFusionClusterList contains a list of TensorFusionCluster.

TensorFusionClusterSpec defines the desired state of TensorFusionCluster.

TensorFusionClusterStatus defines the observed state of TensorFusionCluster.

TensorFusionConnection is the Schema for the tensorfusionconnections API.

TensorFusionConnectionList contains a list of TensorFusionConnection.

TensorFusionConnectionSpec defines the desired state of TensorFusionConnection.

TensorFusionConnectionStatus defines the observed state of TensorFusionConnection.

Use existing Kubernetes GPU nodes.

TensorFusionClusterPhase represents the phase of the TensorFusionCluster resource.