API Reference
Packages
batch.tensorstack.dev/v1beta1
Package v1beta1 defines the CRD types.
Resource Types
- CronWorkflowRun
- CronWorkflowRunList
- WorkflowRun
- WorkflowRunList
- WorkflowTemplate
- WorkflowTemplateList
BeamJobWorkflowTemplate
BeamJobWorkflowTemplate creates a t9k beam job.
Appears in:
| Field | Description |
|---|---|
metadata ObjectMeta | Refer to Kubernetes API documentation for fields of metadata. |
spec BeamJobSpec | Spec of the beam job. |
ColossalAIJobWorkflowTemplate
ColossalAIJobWorkflowTemplate creates a t9k colossalai job.
Appears in:
| Field | Description |
|---|---|
metadata ObjectMeta | Refer to Kubernetes API documentation for fields of metadata. |
spec ColossalAIJobSpec | Spec of the colossalai job. |
ConcurrencyPolicy
Underlying type: string
ConcurrencyPolicy describes how the WorkflowRun will be handled. Only one of the following concurrent policies may be specified. If none of the following policies is specified, the default one is AllowConcurrent.
Appears in:
ConditionSelector
ConditionSelector selects a k8s resource if its status.conditions contains a condition whose type and status are exactly the same as those of the condition selector.
Appears in:
| Field | Description |
|---|---|
type string | Type of the condition. |
status ConditionStatus | Status of the condition, one of True, False, or Unknown. |
CronWorkflowRun
CronWorkflowRun is the Schema for the CronWorkflowRun API
Appears in:
| Field | Description |
|---|---|
apiVersion string | batch.tensorstack.dev/apis |
kind string | CronWorkflowRun |
metadata ObjectMeta | Refer to Kubernetes API documentation for fields of metadata. |
spec CronWorkflowRunSpec | |
status CronWorkflowRunStatus |
CronWorkflowRunCondition
CronWorkflowRunCondition is an observation of the condition of the CronWorkflowRun.
Appears in:
| Field | Description |
|---|---|
type CronWorkflowRunConditionType | Type of CronWorkflowRun condition. |
status ConditionStatus | Status of the condition, one of True, False, or Unknown. |
reason string | The reason for the condition's last transition. |
message string | A readable message indicating details about the transition. |
lastTransitionTime Time | Last time the condition transitioned from one status to another. |
CronWorkflowRunConditionType
Underlying type: string
CronWorkflowRunConditionType defines all possible types of CronWorkflowRunCondition. Can be one of: HasStarted, IsRunning, HasSuccessfulRun, HasFailedRun.
Appears in:
CronWorkflowRunList
CronWorkflowRunList contains a list of CronWorkflowRun
| Field | Description |
|---|---|
apiVersion string | batch.tensorstack.dev/apis |
kind string | CronWorkflowRunList |
metadata ListMeta | Refer to Kubernetes API documentation for fields of metadata. |
items CronWorkflowRun array |
CronWorkflowRunSpec
CronWorkflowRunSpec defines the desired state of CronWorkflowRun.
Appears in:
| Field | Description |
|---|---|
schedule string | Schedule defines the schedule for the running of WorkflowRuns. |
concurrencyPolicy ConcurrencyPolicy | Specifies how to treat concurrent executions of a WorkflowRun. Valid values are: - "Allow" (default): allows WorkflowRuns to run concurrently; - "Forbid": forbids concurrent runs, skipping next run if previous run hasn't finished yet; - "Replace": cancels currently running WorkflowRun and replaces it with a new one |
successfulRunsHistoryLimit integer | The number of successful finished WorkflowRuns to retain. This is a pointer to distinguish between explicit zero and not specified. |
failedRunsHistoryLimit integer | The number of failed finished WorkflowRuns to retain. This is a pointer to distinguish between explicit zero and not specified. |
startingDeadlineSeconds integer | Deadline in seconds for starting the WorkflowRuns if it misses scheduled time for any reason. Missed WorkflowRun runs will be counted as failed ones. |
suspend boolean | This flag tells the controller to suspend subsequent runs, it does not apply to already started runs. Defaults to false. |
workflowRun WorkflowRunTemplateSpec | Specifies the WorkflowRun that will be created when executing a CronWorkflowRun. |
CronWorkflowRunStatus
CronWorkflowRunStatus defines the observed state of CronWorkflowRun.
Appears in:
| Field | Description |
|---|---|
active ObjectReference array | A list of pointers to currently running WorkflowRuns. |
lastScheduleTime Time | The last time when the WorkflowRun was successfully created. |
conditions CronWorkflowRunCondition array | The latest available observations of the CronWorkflowRun's current state. |
DAGNode
DAGNode represents a node in the graph during DAG execution.
Appears in:
| Field | Description |
|---|---|
name string | Name is the name of this node within the DAG. |
workflowTemplateRef string | WorkflowTemplateRef is a reference to a workflowTemplate definition. |
when WhenExpression array | WhenExpressions is a list of when expressions that need to be true for the node to run |
retries integer | Retries represents how many times this node should be retried in case of failure: Condition Succeeded set to False |
dependencies string array | Dependencies are name of other nodes which this depends on, to force a specific ordering in graph execution. |
params Param array | Parameters declares parameters passed to this node. |
workspaces WorkspaceDAGBinding array | Workspaces maps workspaces from the DAG WorkflowTemplate spec to the workspaces declared in the node. |
DAGWorkflowTemplate
DAGWorkflowTemplate creates a dag consisting of other WorkflowTemplates.
Appears in:
| Field | Description |
|---|---|
templates DAGNode array | Templates are a list of DAG nodes. |
failureStrategy FailureStrategy | FailureStrategy is the strategy when a node in DAG fails. |
DeepSpeedJobWorkflowTemplate
DeepSpeedJobWorkflowTemplate creates a t9k deepspeed job.
Appears in:
| Field | Description |
|---|---|
metadata ObjectMeta | Refer to Kubernetes API documentation for fields of metadata. |
spec DeepSpeedJobSpec | Spec of the deepspeed job. |
FailureStrategy
Underlying type: string
FailureStrategy defines the failure strategy of DAGWorkflowTemplate.
Appears in:
GenericJobWorkflowTemplate
GenericJobWorkflowTemplate creates a t9k generic job.
Appears in:
| Field | Description |
|---|---|
metadata ObjectMeta | Refer to Kubernetes API documentation for fields of metadata. |
spec GenericJobSpec | Spec of the generic job. |
MPIJobWorkflowTemplate
MPIJobWorkflowTemplate creates a t9k mpi job.
Appears in:
| Field | Description |
|---|---|
metadata ObjectMeta | Refer to Kubernetes API documentation for fields of metadata. |
spec MPIJobSpec | Spec of the mpi job. |
NodePhase
Underlying type: string
NodePhase is the current state of a node. Node means the node in a DAG. PodWorkflowTemplate, SeqPodWorkflowTemplate and ResourceWorkflowTemplate are also considered as an independent node. DAGWorkflowTemplate itself is considered as a parent node.
Appears in:
NodeStatus
NodeStatus is the status of a node in DAG
Appears in:
| Field | Description |
|---|---|
workflowRunName string | WorkflowRunName is the name of WorkflowRun created from the node. |
phase NodePhase | Phase is the phase of the node. |
whenExpressions WhenExpression array | WhenExpressions show why if the node is skipped. |
Param
Param declares a string to use for the parameter called name.
Appears in:
| Field | Description |
|---|---|
name string | |
value string |
ParamSpec
ParamSpec defines values that are provided by users as inputs on a WorkflowRun.
Appears in:
| Field | Description |
|---|---|
name string | Name declares the name by which a parameter is referenced. |
description string | Description is a user-facing description of the parameter that may be used to populate a UI. |
default string | Default is the value a parameter takes if no input value is supplied. If default is set, a WorkflowTemplate may be executed without a supplied value for the parameter. It serves as a pointer to distinguish between setting no default and setting empty string as default. |
PodSpecWithoutContainers
PodSpecWithoutContainers is a helper struct for SeqPodWorkflowTemplate forked from https://github.com/kubernetes/api/blob/v0.20.0/core/v1/types.go#L2914
Appears in:
| Field | Description |
|---|---|
volumes Volume array | List of volumes that can be mounted by containers belonging to the pod. More info: https://kubernetes.io/docs/concepts/storage/volumes |
initContainers Container array | List of initialization containers belonging to the pod. Init containers are executed in order prior to containers being started. If any init container fails, the pod is considered to have failed and is handled according to its restartPolicy. The name for an init container or normal container must be unique among all containers. Init containers may not have Lifecycle actions, Readiness probes, Liveness probes, or Startup probes. The resourceRequirements of an init container are taken into account during scheduling by finding the highest request/limit for each resource type, and then using the max of of that value or the sum of the normal containers. Limits are applied to init containers in a similar fashion. Init containers cannot currently be added or removed. Cannot be updated. More info: https://kubernetes.io/docs/concepts/workloads/pods/init-containers/ |
ephemeralContainers EphemeralContainer array | List of ephemeral containers run in this pod. Ephemeral containers may be run in an existing pod to perform user-initiated actions such as debugging. This list cannot be specified when creating a pod, and it cannot be modified by updating the pod spec. In order to add an ephemeral container to an existing pod, use the pod's ephemeralcontainers subresource. This field is alpha-level and is only honored by servers that enable the EphemeralContainers feature. |
restartPolicy RestartPolicy | Restart policy for all containers within the pod. One of Always, OnFailure, Never. Default to Always. More info: https://kubernetes.io/docs/concepts/workloads/pods/pod-lifecycle/#restart-policy |
terminationGracePeriodSeconds integer | Optional duration in seconds the pod needs to terminate gracefully. May be decreased in delete request. Value must be non-negative integer. The value zero indicates delete immediately. If this value is nil, the default grace period will be used instead. The grace period is the duration in seconds after the processes running in the pod are sent a termination signal and the time when the processes are forcibly halted with a kill signal. Set this value longer than the expected cleanup time for your process. Defaults to 30 seconds. |
activeDeadlineSeconds integer | Optional duration in seconds the pod may be active on the node relative to StartTime before the system will actively try to mark it failed and kill associated containers. Value must be a positive integer. |
dnsPolicy DNSPolicy | Set DNS policy for the pod. Defaults to "ClusterFirst". Valid values are 'ClusterFirstWithHostNet', 'ClusterFirst', 'Default' or 'None'. DNS parameters given in DNSConfig will be merged with the policy selected with DNSPolicy. To have DNS options set along with hostNetwork, you have to specify DNS policy explicitly to 'ClusterFirstWithHostNet'. |
nodeSelector object (keys:string, values:string) | NodeSelector is a selector which must be true for the pod to fit on a node. Selector which must match a node's labels for the pod to be scheduled on that node. More info: https://kubernetes.io/docs/concepts/configuration/assign-pod-node/ |
serviceAccountName string | ServiceAccountName is the name of the ServiceAccount to use to run this pod. More info: https://kubernetes.io/docs/tasks/configure-pod-container/configure-service-account/ |
serviceAccount string | DeprecatedServiceAccount is a depreciated alias for ServiceAccountName. Deprecated: Use serviceAccountName instead. |
automountServiceAccountToken boolean | AutomountServiceAccountToken indicates whether a service account token should be automatically mounted. |
nodeName string | NodeName is a request to schedule this pod onto a specific node. If it is non-empty, the scheduler simply schedules this pod onto that node, assuming that it fits resource requirements. |
hostNetwork boolean | Host networking requested for this pod. Use the host's network namespace. If this option is set, the ports that will be used must be specified. Default to false. |
hostPID boolean | Use the host's pid namespace. Optional: Default to false. |
hostIPC boolean | Use the host's ipc namespace. Optional: Default to false. |
shareProcessNamespace boolean | Share a single process namespace between all of the containers in a pod. When this is set containers will be able to view and signal processes from other containers in the same pod, and the first process in each container will not be assigned PID 1. HostPID and ShareProcessNamespace cannot both be set. Optional: Default to false. |
securityContext PodSecurityContext | SecurityContext holds pod-level security attributes and common container settings. Optional: Defaults to empty. See type description for default values of each field. |
imagePullSecrets LocalObjectReference array | ImagePullSecrets is an optional list of references to secrets in the same namespace to use for pulling any of the images used by this PodSpec. If specified, these secrets will be passed to individual puller implementations for them to use. For example, in the case of docker, only DockerConfig type secrets are honored. More info: https://kubernetes.io/docs/concepts/containers/images#specifying-imagepullsecrets-on-a-pod |
hostname string | Specifies the hostname of the Pod If not specified, the pod's hostname will be set to a system-defined value. |
subdomain string | If specified, the fully qualified Pod hostname will be " |
affinity Affinity | If specified, the pod's scheduling constraints |
schedulerName string | If specified, the pod will be dispatched by specified scheduler. If not specified, the pod will be dispatched by default scheduler. |
tolerations Toleration array | If specified, the pod's tolerations. |
hostAliases HostAlias array | HostAliases is an optional list of hosts and IPs that will be injected into the pod's hosts file if specified. This is only valid for non-hostNetwork pods. |
priorityClassName string | If specified, indicates the pod's priority. "system-node-critical" and "system-cluster-critical" are two special keywords which indicate the highest priorities with the former being the highest priority. Any other name must be defined by creating a PriorityClass object with that name. If not specified, the pod priority will be default or zero if there is no default. |
priority integer | The priority value. Various system components use this field to find the priority of the pod. When Priority Admission Controller is enabled, it prevents users from setting this field. The admission controller populates this field from PriorityClassName. The higher the value, the higher the priority. |
dnsConfig PodDNSConfig | Specifies the DNS parameters of a pod. Parameters specified here will be merged to the generated DNS configuration based on DNSPolicy. |
readinessGates PodReadinessGate array | If specified, all readiness gates will be evaluated for pod readiness. A pod is ready when all its containers are ready AND all conditions specified in the readiness gates have status equal to "True" More info: https://git.k8s.io/enhancements/keps/sig-network/0007-pod-ready%2B%2B.md |
runtimeClassName string | RuntimeClassName refers to a RuntimeClass object in the node.k8s.io group, which should be used to run this pod. If no RuntimeClass resource matches the named class, the pod will not be run. If unset or empty, the "legacy" RuntimeClass will be used, which is an implicit class with an empty definition that uses the default runtime handler. More info: https://git.k8s.io/enhancements/keps/sig-node/runtime-class.md This is a beta feature as of Kubernetes v1.14. |
enableServiceLinks boolean | EnableServiceLinks indicates whether information about services should be injected into pod's environment variables, matching the syntax of Docker links. Optional: Defaults to true. |
preemptionPolicy PreemptionPolicy | PreemptionPolicy is the Policy for preempting pods with lower priority. One of Never, PreemptLowerPriority. Defaults to PreemptLowerPriority if unset. This field is beta-level, gated by the NonPreemptingPriority feature-gate. |
overhead object (keys:ResourceName, values:Quantity) | Overhead represents the resource overhead associated with running a pod for a given RuntimeClass. This field will be autopopulated at admission time by the RuntimeClass admission controller. If the RuntimeClass admission controller is enabled, overhead must not be set in Pod create requests. The RuntimeClass admission controller will reject Pod create requests which have the overhead already set. If RuntimeClass is configured and selected in the PodSpec, Overhead will be set to the value defined in the corresponding RuntimeClass, otherwise it will remain unset and treated as zero. More info: https://git.k8s.io/enhancements/keps/sig-node/20190226-pod-overhead.md This field is alpha-level as of Kubernetes v1.16, and is only honored by servers that enable the PodOverhead feature. |
topologySpreadConstraints TopologySpreadConstraint array | TopologySpreadConstraints describes how a group of pods ought to spread across topology domains. Scheduler will schedule pods in a way which abides by the constraints. All topologySpreadConstraints are ANDed. |
setHostnameAsFQDN boolean | If true the pod's hostname will be configured as the pod's FQDN, rather than the leaf name (the default). In Linux containers, this means setting the FQDN in the hostname field of the kernel (the nodename field of struct utsname). In Windows containers, this means setting the registry value of hostname for the registry key HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Services\Tcpip\Parameters to FQDN. If a pod does not have FQDN, this has no effect. Default to false. |
PodWorkflowTemplate
PodWorkflowTemplate creates a pod.
Appears in:
| Field | Description |
|---|---|
PodSpec PodSpec |
PyTorchTrainingJobWorkflowTemplate
PyTorchTrainingJobWorkflowTemplate creates a t9k pytorch training job.
Appears in:
| Field | Description |
|---|---|
metadata ObjectMeta | Refer to Kubernetes API documentation for fields of metadata. |
spec PyTorchTrainingJobSpec | Spec of the pytorch training job. |
ResourceWorkflowTemplate
ResourceWorkflowTemplate creates a k8s resource.
Appears in:
| Field | Description |
|---|---|
successRules Rules | SuccessRules is an expression which describes the state of the k8s resource in which the task was considered succeeded. |
failureRules Rules | FailureRules is an expression which describes the state of the k8s resource in which the task was considered failed. |
manifest string | Manifest is the full spec of the k8s resource to create. |
Result
Result used to describe the results of a task
Appears in:
| Field | Description |
|---|---|
name string | Name the given name |
description string | Description is a human-readable description of the result |
value string | Value is a expression to generate DAG result by combining node results Only required for DAG WorkflowTemplate |
Rules
Rules describe the state of the k8s resource.
Appears in:
| Field | Description |
|---|---|
fieldSelector string | FieldSelector selects k8s resources based on the value of one or more resource fields, examples: metadata.name=my-service metadata.namespace!=default status.phase!=Running,spec.restartPolicy=Always |
conditionSelector ConditionSelector | ConditionSelector selects k8s resources based on |
SchedulePolicy
SchedulePolicy defines how k8s schedules the WorkflowRun.
Appears in:
| Field | Description |
|---|---|
t9kScheduler T9kScheduler | T9k Scheduler. |
SeqPodWorkflowTemplate
SeqPodWorkflowTemplate creates a pod whose containers run sequentially. The spec of SeqPodWorkflowTemplate is almost the same with corev1.PodSpec, except that the field Containers is replaced by Steps, to emphasize the sequential execution pattern.
Appears in:
| Field | Description |
|---|---|
PodSpecWithoutContainers PodSpecWithoutContainers | |
steps Step array | Steps are the sequentially running containers of the SeqPodWorkflowTemplate |
Step
Step embeds the Container type, which allows it to include fields not provided by Container.
Appears in:
| Field | Description |
|---|---|
Container Container | |
script string | Script is the contents of an executable file to execute. If Script is not empty, the Step cannot have an Command and the Args will be passed to the Script. |
T9kScheduler
T9kScheduler provides additonal configurations needed for the scheduling process.
Appears in:
| Field | Description |
|---|---|
queue string | Specifies the name of the queue should be used for running this workload. |
TensorFlowTrainingJobWorkflowTemplate
TensorFlowTrainingJobWorkflowTemplate creates a t9k training job.
Appears in:
| Field | Description |
|---|---|
metadata ObjectMeta | Refer to Kubernetes API documentation for fields of metadata. |
spec TensorFlowTrainingJobSpec | Spec of the training job. |
WhenExpression
WhenExpression allows a node in DAG to declare expressions to be evaluated before the node is run to determine whether the node should be executed or skipped.
Appears in:
| Field | Description |
|---|---|
input string | Input is the string for guard checking which can be a static input or an output from a parent node |
operator Operator | Operator that represents an Input's relationship to the values |
values string array | Values is an array of strings, which is compared against the input, for guard checking It must be non-empty |
WorkflowRun
WorkflowRun is the Schema for the workflowrun API
Appears in:
| Field | Description |
|---|---|
apiVersion string | batch.tensorstack.dev/apis |
kind string | WorkflowRun |
metadata ObjectMeta | Refer to Kubernetes API documentation for fields of metadata. |
spec WorkflowRunSpec | |
status WorkflowRunStatus |
WorkflowRunCondition
WorkflowRunCondition is an observation of the condition of the WorkflowRun.
Appears in:
| Field | Description |
|---|---|
type WorkflowRunConditionType | Type of WorkflowRun condition. |
status ConditionStatus | Status of the condition, one of True, False, or Unknown. |
reason string | The reason for the condition's last transition. |
message string | A readable message indicating details about the transition. |
lastTransitionTime Time | Last time the condition transitioned from one status to another. |
WorkflowRunConditionType
Underlying type: string
WorkflowRunConditionType defines all possible types of WorkflowRunConditionType. Can be one of: Running, Completed.
Appears in:
WorkflowRunList
WorkflowRunList contains a list of WorkflowRun
| Field | Description |
|---|---|
apiVersion string | batch.tensorstack.dev/apis |
kind string | WorkflowRunList |
metadata ListMeta | Refer to Kubernetes API documentation for fields of metadata. |
items WorkflowRun array |
WorkflowRunResult
WorkflowRunResult used to describe the results of a workflowTemplate
Appears in:
| Field | Description |
|---|---|
name string | Name the given name |
value string | Value the given value of the result |
WorkflowRunSpec
WorkflowRunSpec defines the desired state of WorkflowRun
Appears in:
| Field | Description |
|---|---|
workspaces WorkspaceBinding array | Workspaces is a list of mappings from volumes to workspace names that will be supplied to WorkflowTemplate. |
params Param array | Params is a list of input parameters that will be supplied to WorkflowTemplate. |
serviceAccountName string | ServiceAccountName is the name of the service account that the created pod will use. |
workflowTemplateRef string | WorkflowTemplateRef is the name of the WorkflowTemplate that WorkflowRun will use. No more than one of the WorkflowTemplateRef and WorkflowTemplateSpec may be specified. |
workflowTemplateSpec WorkflowTemplateSpec | WorkflowTemplateSpec is an embedded spec of WorkflowTemplate that WorkflowRun will use. No more than one of the WorkflowTemplateRef and WorkflowTemplateSpec may be specified. |
timeout Duration | Time after which the build times out. Defaults to never. Refer Go's ParseDuration documentation for expected format: https://golang.org/pkg/time/#ParseDuration |
depth integer | Depth indicates the depth of the WorkflowRun in DAG. If Depth exceeds limit, the WorkflowRun will not be created. |
scheduler SchedulePolicy | Identifies the preferred scheduler for allocating resources to replicas. Defaults to cluster default scheduler. |
WorkflowRunStatus
WorkflowRunStatus defines the observed state of WorkflowRun
Appears in:
| Field | Description |
|---|---|
WorkflowRunStatusFields WorkflowRunStatusFields | WorkflowRunStatusFields inlines the status fields. |
retriedStatus WorkflowRunStatusFields array | RetriedStatus contains the history of WorkflowRunStatus in case of a retry in order to keep record of failures. |
WorkflowRunStatusFields
WorkflowRunStatusFields are the main fields of WorkflowRunStatus
Appears in:
| Field | Description |
|---|---|
phase NodePhase | Phase is a simple, high-level summary of where the WorkflowRun is in its lifecycle. |
message string | A human readable message indicating details about why the WorkflowRun is in this condition. |
nodes object (keys:string, values:NodeStatus) | Nodes is a map of DAG nodes status, with the node name as the key. |
startTime Time | StartTime is the time the build is actually started. |
completionTime Time | CompletionTime is the time the build completed. |
results WorkflowRunResult array | Results are the list of results written out by the workflowTemplate's containers |
workflowTemplateSpec WorkflowTemplateSpec | WorkflowTemplateSpec contains the Spec from the dereferenced WorkflowTemplate definition used to instantiate this WorkflowRun. |
podName string | PodName is the name of the pod created by WorkflowRun. |
conditions WorkflowRunCondition array | The latest available observations of the WorkflowRun's current state. |
WorkflowRunTemplateSpec
WorkflowRunTemplateSpec describes the data a WorkflowRun should have when created from a template
Appears in:
| Field | Description |
|---|---|
metadata ObjectMeta | Refer to Kubernetes API documentation for fields of metadata. |
spec WorkflowRunSpec | Specification of the desired behavior of the WorkflowRun. |
WorkflowTemplate
WorkflowTemplate is the Schema for the WorkflowTemplate API
Appears in:
| Field | Description |
|---|---|
apiVersion string | batch.tensorstack.dev/apis |
kind string | WorkflowTemplate |
metadata ObjectMeta | Refer to Kubernetes API documentation for fields of metadata. |
spec WorkflowTemplateSpec | |
status WorkflowTemplateStatus |
WorkflowTemplateList
WorkflowTemplateList contains a list of WorkflowTemplate
| Field | Description |
|---|---|
apiVersion string | batch.tensorstack.dev/apis |
kind string | WorkflowTemplateList |
metadata ListMeta | Refer to Kubernetes API documentation for fields of metadata. |
items WorkflowTemplate array |
WorkflowTemplateSpec
WorkflowTemplateSpec defines the desired state of WorkflowTemplate.
Appears in:
| Field | Description |
|---|---|
description string | Description is a user-facing description of the task that may be used to populate a UI. |
workspaces WorkspaceDeclaration array | Workspaces are the volumes that this WorkflowTemplate requires. Workspaces must be supplied as inputs in WorkflowRuns unless they are declared as optional. |
params ParamSpec array | Params is a list of input parameters required to run the task. Params must be supplied as inputs in WorkflowRuns unless they declare a default value. |
results Result array | Results are values that this WorkflowTemplate can output |
type WorkflowTemplateType | Type defines the type of WorkflowTemplate. If empty, the first non-nil field in (pod, seqPod, resource, dag) will be the type. |
pod PodWorkflowTemplate | Pod is the spec of pod which WorkflowTemplate will create. |
seqPod SeqPodWorkflowTemplate | SeqPod is the spec of pod with sequentially running containers which WorkflowTemplate will create. |
resource ResourceWorkflowTemplate | Resource is the spec of k8s resource which WorkflowTemplate will create. |
dag DAGWorkflowTemplate | DAG is the spec of DAG which WorkflowTemplate will create. |
genericJob GenericJobWorkflowTemplate | GenericJob is the spec of t9k generic job which WorkflowTemplate will create. |
tensorflowTrainingJob TensorFlowTrainingJobWorkflowTemplate | TensorFlowTrainingJob is the spec of t9k training job which WorkflowTemplate will create. |
pytorchTrainingJob PyTorchTrainingJobWorkflowTemplate | PyTorchTrainingJob is the spec of t9k pytorch training job which WorkflowTemplate will create. |
xgboostTrainingJob XGBoostTrainingJobWorkflowTemplate | XGBoostTrainingJob is the spec of t9k xgboost training job which WorkflowTemplate will create. |
colossalaiJob ColossalAIJobWorkflowTemplate | ColossalAIJob is the spec of t9k colossalai job which WorkflowTemplate will create. |
deepspeedJob DeepSpeedJobWorkflowTemplate | DeepSpeedJob is the spec of t9k deepspeed job which WorkflowTemplate will create. |
mpiJob MPIJobWorkflowTemplate | MPIJob is the spec of t9k mpi job which WorkflowTemplate will create. |
beamJob BeamJobWorkflowTemplate | BeamJob is the spec of t9k beam job which WorkflowTemplate will create. |
WorkflowTemplateType
Underlying type: string
WorkflowTemplateType defines the type of WorkflowTemplate.
Appears in:
WorkspaceBinding
WorkspaceBinding maps a WorkflowTemplate's declared workspace to a Volume.
Appears in:
| Field | Description |
|---|---|
name string | Name is the name of the workspace populated by the volume. |
subPath string | SubPath is optionally a directory on the volume which should be used for this binding (i.e. the volume will be mounted at this sub directory). |
persistentVolumeClaim PersistentVolumeClaimVolumeSource | PersistentVolumeClaimVolumeSource represents a reference to a PersistentVolumeClaim in the same namespace. Either this OR EmptyDir can be used. |
emptyDir EmptyDirVolumeSource | EmptyDir represents a temporary directory that shares a WorkflowTemplate's lifetime. More info: https://kubernetes.io/docs/concepts/storage/volumes#emptydir Either this OR PersistentVolumeClaim can be used. |
configMap ConfigMapVolumeSource | ConfigMap represents a configMap that should populate this workspace. |
secret SecretVolumeSource | Secret represents a secret that should populate this workspace. |
WorkspaceDAGBinding
WorkspaceDAGBinding describes how a workspace passed into a DAG WorkflowTemplate should be mapped to a node's declared workspace.
Appears in:
| Field | Description |
|---|---|
name string | Name is the name of the workspace as declared by the node |
workspace string | Workspace is the name of the workspace declared by the DAG WorkflowTemplate |
subPath string | SubPath is optionally a directory on the volume which should be used for this binding (i.e. the volume will be mounted at this sub directory). |
WorkspaceDeclaration
WorkspaceDeclaration is a declaration of a volume that a WorkflowTemplate requires.
Appears in:
| Field | Description |
|---|---|
name string | Name is the name by which you can bind the volume at runtime. |
description string | Description is an optional human readable description of this volume. |
mountPath string | MountPath overrides the directory that the volume will be made available at. |
readOnly boolean | ReadOnly dictates whether a mounted volume is writable. By default this field is false and so mounted volumes are writable. |
optional boolean | Optional marks a Workspace as not being required in WorkflowRuns. By default this field is false and so declared workspaces are required. |
XGBoostTrainingJobWorkflowTemplate
XGBoostTrainingJobWorkflowTemplate creates a t9k xgboost training job.
Appears in:
| Field | Description |
|---|---|
metadata ObjectMeta | Refer to Kubernetes API documentation for fields of metadata. |
spec XGBoostTrainingJobSpec | Spec of the xgboost training job. |