Additional Configuration

Configuration

This document outlines the configurable parameters for the Cedana Helm chart, as defined in the values.yaml file. For up to date configuration, see https://github.com/cedana/cedana-helm-charts/blob/main/cedana-helm/values.yaml.

Global Settings

These settings control the overall behavior of the deployment.

Parameter	Description	Default
nameOverride	Overrides the name of the chart.	"cedana"
fullnameOverride	Overrides the full name of the release.	"cedana"
installKueue	If set to true, Kueue will be installed. Note: The Kueue CRDs must be applied before enabling this option. You can apply them with kubectl apply --server-side -f https://github.com/kubernetes-sigs/kueue/releases/download/v0.10.1/manifests.yaml.	false
kubernetesClusterDomain	The Kubernetes cluster domain.	cluster.local

Cedana Configuration (cedanaConfig)

This section contains the core configuration for the Cedana platform.

Parameter	Description	Default
cedanaAuthToken	Your authentication token for the Cedana platform.	""
cedanaUrl	The URL for the Cedana API.	""
cedanaClusterName	A unique name for your cluster within the Cedana platform.	""
cedanaSqsQueueUrl	The SQS queue URL for communication with Cedana.	""
checkpointStreams	The number of parallel streams for checkpoint/restore operations. 0 disables streaming.	4
checkpointCompression	The compression algorithm for checkpoints. Options: "none", "tar", "lz4", "gzip", "zlib".	"lz4"
gpuPoolSize	The number of GPU controllers to keep warm to improve startup/restore time for GPU workloads.	0
gpuFreezeType	The default freeze type for GPU workloads. Options: "IPC", "NCCL".	"IPC"
gpuShmSize	The shared memory size for GPU workloads. It is set to 8 GiB by default.	"8589934592"
gpuLdLibPath	Additional LD_LIBRARY_PATH to locate CUDA libraries.	"/run/nvidia/driver/usr/lib/x86_64-linux-gnu"
pluginsBuilds	Specifies the build type for plugins. Use "release" for stable versions or "alpha" for development branches.	"release"
pluginsNativeVersion	The version of the native plugin to use.	"latest"
pluginsCriuVersion	The version of the CRIU plugin to use.	"v4.1-cedana.01"
pluginsRuntimeShimVersion	The version of the runtime shim plugin to use.	"v0.6.1"
pluginsGpuVersion	The version of the GPU plugin to use.	"v0.5.5"
pluginsStreamerVersion	The version of the streamer plugin to use.	"v0.0.8"
profilingEnabled	If true, enables profiling.	true
metricsOtel	If true, enables OpenTelemetry metrics.	true
logLevel	The logging level.	"info"
preExistingSecret	If you want to use a custom pre-existing secret, uncomment this and provide the name.	cedana-secret-user (commented out)

Daemon Helper (daemonHelper)

Configuration for the daemon-helper DaemonSet.

Parameter	Description	Default
upgradeAndRestart	If true, the daemon helper will be upgraded and restarted.	false
service.annotations	Annotations to add to the daemon helper service.	{}
image.repository	The repository for the cedana-helper image.	cedana/cedana-helper
image.tag	The tag for the cedana-helper image.	v0.9.251
image.imagePullPolicy	The image pull policy.	IfNotPresent
updateStrategy.maxSurge	The maximum number of pods that can be created over the desired number of pods.	0
updateStrategy.maxUnavailable	The maximum number of pods that can be unavailable during the update process.	1
tolerations	Tolerations for the daemon helper pods.	[]
affinity	Affinity settings for the daemon helper pods.	{}

Service Account (serviceAccount)

Configuration for the Kubernetes Service Account.

Parameter	Description	Default
create	If true, a Service Account will be created.	true
automount	If true, a ServiceAccount's API credentials will be automatically mounted.	true
annotations	Annotations to add to the Service Account.	{}
name	The name of the Service Account. If not set and create is true, a name is generated.	"cedana-controller-manager"

Controller Manager (controllerManager)

Configuration for the cedana-controller-manager.

Parameter	Description	Default
autoscaling.enabled	If true, enables autoscaling for the controller manager.	false
autoscaling.replicaCount	The number of replicas for the controller manager.	1
autoscaling.deploymentRevisionHistoryLimit	The number of old ReplicaSets to retain.	10
service.annotations	Annotations for the controller manager service.	{}
service.ports	The ports for the controller manager service.	TCP/1324
manager.podAnnotations	Annotations for the controller manager pods.	{}
manager.args	Arguments for the controller manager container.	[--health-probe-bind-address=:8081, --metrics-bind-address=127.0.0.1:8080, --leader-elect]
manager.containerSecurityContext	The security context for the manager container, which is configured to be non-privileged.	allowPrivilegeEscalation: false, capabilities: { drop: [ALL] }
manager.image.repository	The repository for the cedana-controller image.	cedana/cedana-controller
manager.image.tag	The tag for the cedana-controller image.	v0.4.5
manager.image.imagePullPolicy	The image pull policy.	IfNotPresent
manager.resources	Resource limits and requests for the manager container. Default is minimal. Uncomment to set custom values.	{}
rbac.resources	Resource limits and requests for the RBAC proxy container. Default is minimal. Uncomment to set custom values.	{}
tolerations	Tolerations for the controller manager pods.	[]
affinity	Affinity settings for the controller manager pods.	{}

Shared Memory Configuration (shmConfig)

Optional configuration to increase /dev/shm size on nodes, which is useful for workloads requiring large shared memory.

Parameter	Description	Default
enabled	Set to true to enable the increase of /dev/shm size.	false
size	The desired size for /dev/shm (e.g., "10G", "20G").	"10G"
minBytes	The minimum size in bytes that will trigger a remount of /dev/shm.	"10737418240"

Metrics Service (metricsService)

Configuration for the metrics service.

Parameter	Description	Default
ports	The ports for the metrics service.	HTTPS/8443
type	The type of the metrics service.	ClusterIP