Configure auto scaling settings for a cluster with SetAutoScaleConfig - Elastic High Performance Computing - Alibaba Cloud - Elastic High Performance Computing

Configures auto scaling for a cluster.

Operation description

Usage notes

Queue-level settings override global settings.

Try it now

Try this API in OpenAPI Explorer, no manual signing needed. Successful calls auto-generate SDK code matching your parameters. Download it with built-in credential security for local usage.

Test

RAM authorization

The table below describes the authorization required to call this API. You can define it in a Resource Access Management (RAM) policy. The table's columns are detailed below:

Action: The actions can be used in the Action element of RAM permission policy statements to grant permissions to perform the operation.
API: The API that you can call to perform the action.
Access level: The predefined level of access granted for each API. Valid values: create, list, get, update, and delete.
Resource type: The type of the resource that supports authorization to perform the action. It indicates if the action supports resource-level permission. The specified resource must be compatible with the action. Otherwise, the policy will be ineffective.
- For APIs with resource-level permissions, required resource types are marked with an asterisk (*). Specify the corresponding Alibaba Cloud Resource Name (ARN) in the Resource element of the policy.
- For APIs without resource-level permissions, it is shown as All Resources. Use an asterisk (*) in the Resource element of the policy.
Condition key: The condition keys defined by the service. The key allows for granular control, applying to either actions alone or actions associated with specific resources. In addition to service-specific condition keys, Alibaba Cloud provides a set of common condition keys applicable across all RAM-supported services.
Dependent action: The dependent actions required to run the action. To complete the action, the RAM user or the RAM role must have the permissions to perform all dependent actions.

Action

Access level

Resource type

Condition key

Dependent action

ehpc:SetAutoScaleConfig

*All Resource

*

None

Request parameters

Parameter	Type	Required	Description	Example
ClusterId	string	Yes	The cluster ID.	ehpc-hz-FYUr32****
EnableAutoGrow	boolean	No	Specifies whether to enable auto scale-out. Valid values: true: enables auto scale-out. false: disables auto scale-out. Default value: false.	false
EnableAutoShrink	boolean	No	Specifies whether to enable auto scale-in. Valid values: true: enables auto scale-in. false: disables auto scale-in. Default value: false.	false
GrowIntervalInMinutes	integer	No	The interval between two consecutive rounds of scale-out. Unit: minutes. Valid values: 2 to 10. Default value: 2. Note A gap may occur between rounds within a scale-out task or between consecutive tasks.	2
ShrinkIntervalInMinutes	integer	No	The interval between two consecutive rounds of scale-in. Unit: minutes. Valid values: 2 to 10. Default value: 2.	2
ShrinkIdleTimes	integer	No	The number of consecutive idle checks before a compute node is released. Valid values: 2 to 5. Default value: 3. With the default values (ShrinkIdleTimes=3, ShrinkIntervalInMinutes=2), a node idle for more than 6 consecutive minutes is released.	3
GrowTimeoutInMinutes	integer	No	The scale-out timeout period. Unit: minutes. Valid values: 10 to 60. Default value: 20. Nodes that do not reach the Running state within this period are released.	20
ExtraNodesGrowRatio	integer	No	The percentage of extra compute nodes. Valid values: 0 to 100. Default value: 0. If you need to add 100 compute nodes and the value of the ExtraNodesGrowRatio parameter is 2, 102 compute nodes are added.	0
GrowRatio	integer	No	The percentage of each round of a scale-out task. Valid values: 1 to 100. Default value: 100. If you set GrowRatio to 50, the scale-out has two rounds. Each round completes half of the scale-out.	50
MaxNodesInCluster	integer	Yes	The maximum number of compute nodes that can be added to the cluster. Valid values: 0 to 500. Default value: 100.	100
ExcludeNodes	string	No	The compute nodes excluded from auto scaling. Separate multiple nodes with commas (,). Specified nodes are retained even when idle.	i-bp19lgqwxb4206t5**,i-bp1g4hvzs9pywrhb**
SpotStrategy	string	No	The bidding policy for compute nodes. Valid values: NoSpot: pay-as-you-go instances. SpotWithPriceLimit: preemptible instances with a user-defined maximum hourly price. SpotAsPriceGo: preemptible instances at the current market price. Default value: NoSpot.	SpotWithPriceLimit
SpotPriceLimit	number	No	The maximum hourly price of compute nodes. Accurate to three decimal places. Takes effect only when `SpotStrategy` is set to `SpotWithPriceLimit`.	0.062
ImageId	string	No	The IDs of the images. Note If both `Queues.N.QueueImageId` and `ImageId` are specified, `Queues.N.QueueImageId` prevails. If you set `Queues.N.QueueImageId` or `ImageId`, the parameter that you set takes effect. If you leave both `Queues.N.QueueImageId` and `ImageId` empty, the image that was specified when you created the cluster or the last time you scaled out the cluster is used by default.	centos_7_03_64_20G_alibase_201708****
Queues	array<object>	No	The information about the queue.
	array<object>	No	The array of information about queues.
QueueImageId	string	No	The image ID of the queue where the scale-out is performed. Valid values of N: 1 to 8. Note If you set both `Queues.N.QueueImageId` and `ImageId`, `Queues.N.QueueImageId` prevails. If you set `Queues.N.QueueImageId` or `ImageId`, the parameter that you set takes effect. If you leave both `Queues.N.QueueImageId` and `ImageId` empty, the image that was specified when you created the cluster or the last time when you scaled out the cluster is used by default.	centos_7_03_64_20G_alibase_201708****
SystemDiskCategory	string	No	The system disk category for auto-added compute nodes. Valid values: cloud_efficiency: ultra disk. cloud_ssd: standard SSD. cloud_essd: enhanced SSD (ESSD). cloud: basic disk. Disks of this type are retired. Valid values of N: 1 to 8. Default value: cloud_efficiency.	cloud_efficiency
InstanceType	string	No	The instance type of auto-added compute nodes. Valid values of N: 1 to 8.	ecs.n1.medium
HostNameSuffix	string	No	The hostname suffix for scale-out nodes in the queue. Valid values of N: 1 to 8.	000
SpotStrategy	string	No	The bidding policy for auto-added compute nodes. Valid values of N: 1 to 8. Valid values: NoSpot: pay-as-you-go instance. SpotWithPriceLimit: preemptible instance with a user-defined maximum hourly price. SpotAsPriceGo: preemptible instance at the current market price. Default value: NoSpot.	NoSpot
HostNamePrefix	string	No	The hostname prefix for scale-out nodes in the queue. Valid values of N: 1 to 8.	compute
MinNodesInQueue	integer	No	The minimum number of compute nodes that can be removed from the queue. Valid values: 0 to 50. Valid values of N: 1 to 8. Default value: 0.	0
SystemDiskSize	integer	No	The system disk size for auto-added compute nodes. Unit: GB. Valid values: 40 to 500. Valid values of N: 1 to 8. Default value: 40.	40
MaxNodesInQueue	integer	No	The maximum number of compute nodes that can be added to the queue. Valid values: 0 to 500. Valid values of N: 1 to 8. Default value: 100.	30
EnableAutoShrink	boolean	No	Specifies whether the queue enables auto scale-in. Valid values: true: enables auto scale-in. false: disables auto scale-in. Valid values of N: 1 to 8. Default value: false.	false
QueueName	string	No	The name of the queue. You can specify 1 to 8 names.	cluster1
EnableAutoGrow	boolean	No	Specifies whether the queue enables auto scale-out. Valid values: true: enables auto scale-out. false: disables auto scale-out. Valid values of N: 1 to 8. Default value: false.	false
SystemDiskLevel	string	No	The system disk performance level for auto-added compute nodes. Valid values: PL0: An ESSD can deliver up to 10,000 random read/write IOPS. PL1: An ESSD can deliver up to 50,000 random read/write IOPS. PL2: An ESSD can deliver up to 100,000 random read/write IOPS. PL3: An ESSD can deliver up to 1,000,000 random read/write IOPS. Valid values of N: 1 to 8. Default value: PL1.	PL1
SpotPriceLimit	number	No	The maximum hourly price of auto-added compute nodes. Accurate to three decimal places. Takes effect only when `Queues.N.SpotStrategy` is set to `SpotWithPriceLimit`. Valid values of N: 1 to 8.	0.662
InstanceTypes	array<object>	No	The instance types of the nodes in the queue.
	object	No	The array of node information.
VSwitchId	string	No	The vSwitch ID for auto-added compute nodes in the queue. Valid values of N: 0 to 500.	vsw-bp1lfcjbfb099rrjn****
ZoneId	string	No	The zone ID for auto-added compute nodes in the queue. Valid values of N: 0 to 500.	cn-hangzhou-b
SpotPriceLimit	number	No	The maximum hourly price of auto-added compute nodes. Accurate to three decimal places. Takes effect only when `Queues.N.InstanceTypes.N.SpotStrategy` is set to `SpotWithPriceLimit`. Valid values of N: 0 to 500.	0.660
InstanceType	string	No	The instance type of auto-added compute nodes in the queue. Valid values of N: 0 to 500.	ecs.n1.tiny
SpotStrategy	string	No	The bidding policy for auto-added compute nodes. Valid values: NoSpot: pay-as-you-go instance. SpotWithPriceLimit: preemptible instance with a user-defined maximum hourly price. SpotAsPriceGo: preemptible instance at the current market price. Default value: NoSpot. Valid values of N: 0 to 500.	SpotWithPriceLimit
SpotDuration	integer	No	The protection period of the preemptible instance. Unit: hours. Valid values: 0 to 1. A value of 0 means that no protection period is specified. Default value: 1.	0
SpotInterruptionBehavior	string	No	The interruption mode of the preemptible instance. The value can only be Terminate, which indicates that the instance is released.	Terminate
DataDisks	array<object>	No	The list of data disks.
	object	No	The list of data disks.
DataDiskSize	integer	No	The size of the data disk. Unit: GB. Valid values: 40 to 500. Default value: 40. Valid values of N: 0 to 16.	40
DataDiskCategory	string	No	The category of the data disk. Valid values: cloud_efficiency: ultra disk. cloud_ssd: standard SSD. cloud_essd: ESSD. cloud: basic disk. Default value: cloud_efficiency. Valid values of N: 0 to 16.	cloud_efficiency
DataDiskDeleteWithInstance	boolean	No	Specifies whether the data disk is released when the node is released. Valid values: true false Default value: true. Valid values of N: 0 to 16.	true
DataDiskPerformanceLevel	string	No	The performance level of the ESSD used as the data disk. The parameter takes effect only when the Queues.N.DataDisks.N.DataDiskCategory parameter is set to cloud_essd. Valid values: PL0: An ESSD can deliver up to 10,000 random read/write IOPS. PL1: An ESSD can deliver up to 50,000 random read/write IOPS. PL2: An ESSD can deliver up to 100,000 random read/write IOPS. PL3: An ESSD can deliver up to 1,000,000 random read/write IOPS. Default value: PL1. Valid values of N: 0 to 16.	PL1
DataDiskEncrypted	boolean	No	Specifies whether to encrypt the data disk. Valid values: true false Default value: false. Valid values of N: 0 to 16.	false
DataDiskKMSKeyId	string	No	The Key Management Service (KMS) key ID of the data disk. Valid values of N: 0 to 16.	0e478b7a-4262-4802-b8cb-00d3fb40826X
MaxNodesPerCycle	integer	No	The maximum number of compute nodes that can be added in each round of scale-out. Valid values: 0 to 99. Default value: 0.	20
MinNodesPerCycle	integer	No	The minimum number of compute nodes added per scale-out round. Valid values: 1 to 99. Default value: 1. If the requested node count is less than this value, the system adjusts automatically to match the request. Note This setting applies only to the current round.	5
AutoMinNodesPerCycle	boolean	No	Whether to automatically set the minimum scale-out count to match the job requirement per round. Maximum: 99.	false
SortedByInventory	boolean	No	Whether to sort instance types by inventory during auto scaling. Valid values: true false Note When true, the system selects instance types in descending order of stock availability.	false
ComputeEnableHt	boolean	No	Specifies whether to enable hyper-threading for the ECS instance that is used as the compute node. Note Hyper-threading is enabled by default and can only be disabled for certain instance types. Specify and view CPU options.	true
DnsConfig	string	No	The DNS configuration in JSON format.	"{\"DnsType\":\"PrivateZone\",\"DnsName\":\"xxxxx\"}"

Response elements

Element	Type	Description	Example
	object
RequestId	string	The request ID.	04F0F334-13358736C-A1D7-6C044FE7****

Examples

Success response

JSON format

{
  "RequestId": "04F0F334-13358736C-A1D7-6C044FE7****"
}

Error codes

HTTP status code	Error code	Error message	Description
400	InvalidParams	The specified parameter %s is invalid.	The specified parameter %s is invalid.
500	UnknownError	An unknown error occurred.	An unknown error occurred.
403	InvalidOpt	The specified operation is invalid. Please try again later. If the problem still exists, submit a ticket.
403	TooManyComputes	The number of computing nodes exceeds the quota.	The number of computing nodes exceeds the quota.
404	ClusterNotFound	The specified cluster does not exist.	The specified instance does not exist.
404	ImageNotFound	The specified image does not exist.	The specified image does not exist. Please verify the parameter.
503	ServiceUnavailable	The request has failed due to a temporary failure of the server	The request has failed due to a temporary failure of the server.
406	DbError	A database service error occurred.	Database request failed.
407	NotAuthorized	You are not authorized by RAM for this request.	The request is not authorized by RAM.

See Error Codes for a complete list.

Release notes

See Release Notes for a complete list.