What's new

Feature

Description

Release date

Region

Related documentation

Lifecycle hooks support EventBridge notifications.

After you configure EventBridge for a lifecycle hook, the system generates different types of lifecycle notifications when a scaling activity occurs in the scaling group. A scale-out lifecycle notification sends an ess:LifecycleHook:ScaleOut event, and a scale-in lifecycle notification sends an ess:LifecycleHook:ScaleIn event. You can use these notifications to identify and handle the corresponding events.

2023-06-26

All

• CreateLifecycleHook - Create one or more lifecycle hooks for a scaling group

• ModifyLifecycleHook - Modify a lifecycle hook

• DescribeLifecycleHooks - Query the details of lifecycle hooks

Feature

Description

Release date

Region

Related documentation

System event-triggered tasks support the Concurrent Connections metric for ECS.

For system event-triggered tasks in an ECS scaling group, you can select the (ECS) ECS Concurrent Connections (ConcurrentConnections) metric. This metric monitors the number of concurrent network connections on an ECS instance to help you better understand its network load.

2023-05-25

All

System event-triggered tasks

Feature

Description

Release date

Region

Related documentation

The health diagnosis feature now includes cost and inventory diagnosis metrics.

Cost diagnosis metrics check whether your account has overdue payments. Inventory diagnosis metrics check the inventory health of the instance types specified in the scaling group configuration. These metrics help you promptly identify and resolve common issues.

2023-04-18

All

Health diagnosis

Auto Scaling supports setting rules for ordered and unique instance names (InstanceName).

You can set rules for the instance name (InstanceName) or hostname (HostName) in a scaling configuration. After a scale-out, Auto Scaling generates ordered and unique instance names or hostnames for the scaled-out ECS instances, which simplifies instance management.

2023-04-15

All

Set rules for ordered and unique instance names or hostnames

You can now log on to ECS instances created by Auto Scaling with a standard account (ecs-user).

When you create an ECS-based scaling configuration via an API call, setting the ImageOptions.LoginAsNonRoot parameter to Yes allows you to log on to instances created during scale-out activities by using a standard account (ecs-user).

2023-04-12

All

CreateScalingConfiguration - Create an ECS-based scaling configuration

Scaling groups support setting weights for backend server groups associated with load balancers.

After you associate a scaling group with a load balancer, you can specify the weight for an instance when it is added to the backend server group of the load balancer. You can specify the weight when you create the scaling group or set the weight parameter in the scaling configuration.

2023-04-10

All

Add or remove load balancers for a scaling group

Auto Scaling supports the single-instance QPS metric for ALB server groups.

You can configure an event-triggered task based on the single-instance QPS metric of an ALB server group to automatically trigger scaling activities and improve the high availability of your applications.

2023-04-03

All

Automatically scale ECI instances based on the single-instance QPS metric of an ALB

Feature

Description

Release date

Region

Related documentation

Auto Scaling supports the health diagnosis feature.

The health diagnosis feature of Auto Scaling performs a comprehensive check on the configuration of a scaling group. This helps you understand the operational status of the scaling group, promptly identify and resolve common issues, and improve the success rate of scale-out and scale-in activities.

2023-03-21

All

Health diagnosis

Feature

Description

Release date

Region

Related documentation

Scaling groups support association with NLB server groups.

After a scaling group is associated with an NLB server group, the ECS or ECI instances that are added to the scaling group are automatically added as backend servers to the NLB server group. This accelerates the processing of access requests distributed by the NLB instance.

2023-02-03

All

• AttachServerGroups

• DetachServerGroups

Feature

Description

Release date

Region

Related documentation

ECI-based scaling configurations support setting the update method for the Containers parameter.

You can update the Containers parameter in ECI-based scaling configurations using either a full or an incremental update. You can choose the appropriate update method based on your business requirements.

2023-01-03

All

ModifyEciScalingConfiguration

Scaling configurations support querying the image source.

You can call the DescribeScalingConfigurations API operation to query the source of the image used in a scaling configuration, such as a public image, custom image, or Alibaba Cloud Marketplace image.

2023-01-10

All

DescribeScalingConfigurations

Feature

Description

Release date

Region

References

Scaling groups support resource groups

You can use resource groups to manage scaling groups. When you create a scaling group, you can assign it to a resource group. This lets you grant resource group-level permissions for fine-grained control over your scaling groups.

2022-12-27

All

• CreateScalingGroup

• ModifyScalingGroup

• DescribeScalingGroups

Feature

Description

Release date

Region

References

Lifecycle hooks can now pass thelifecycleActionResult parameter to OOS templates

When a lifecycle hook pauses a scaling activity, the notification sent to an OOS template now includes thelifecycleActionResult parameter. This parameter specifies the next action to take after the lifecycle hook's wait state ends. You can use this parameter to end the pending activity early if the OOS template execution fails.

2022-11-02

All

• CreateLifecycleHook

• ModifyLifecycleHook

Scaling groups with an expected number of instances support synchronous scaling activities

For scaling groups configured with an expected number of instances, you can set theSyncActivity parameter. This parameter enables synchronous and asynchronous execution modes. You can choose the mode that best fits your business needs to improve scaling activity efficiency.

2022-11-02

All

ScaleWithAdjustment

Feature

Description

Release date

Region

References

Auto Scaling supports custom scale-in policies by using Function Compute

You can select a custom removal policy by setting the RemovalPolicy parameter of a scaling group to CustomPolicy and specifying a value for the CustomPolicyARN parameter. This means that the scaling group removes ECS instances based on a custom scale-in policy (function), which allows you to flexibly select an instance removal method that suits your business scenarios.

2022-10-14

All

• Use Function Compute to create a custom scale-in policy for ECS instances

• CreateScalingGroup

• ModifyScalingGroup

Lifecycle hooks can be enabled or paused

You can enable or pause a lifecycle hook in the Auto Scaling console or by calling the ModifyLifecycleHook API. A paused lifecycle hook does not take effect during scaling activities. Pause or enable this feature based on your specific business scenarios.

2022-10-13

All

• Enable or pause a lifecycle hook

• ModifyLifecycleHook

• DescribeLifecycleHooks

Feature

Description

Release date

Region

References

Scaling groups for ECS instances support setting a maximum instance lifetime

You can set a maximum lifetime for instances in a scaling group. When an ECS instance exceeds its maximum lifetime, Auto Scaling automatically creates a new ECS instance to replace it.

2022-08-25

All

• Create a scaling group

• CreateScalingGroup

• ModifyScalingGroup

• DescribeScalingGroups

Feature

Description

Release date

Region

References

Scaling groups for ECI instances support cost optimization

After you enable the cost optimization feature, the scaling group queries prices, matches specifications, and compares costs before a scale-out. Based on your application's ECI configuration, the scaling group automatically creates ECI instances by using lower-priced ECS specifications. This lets you manage and scale your Elastic Container Instance (ECI) applications more cost-effectively.

2022-07-29

All

• CreateEciScalingConfiguration

• ModifyEciScalingConfiguration

• DescribeEciScalingConfigurations

Event-triggered tasks support multi-metric alert rules

You can configure event-triggered tasks with alert rules that trigger only when multiple metrics meet their thresholds simultaneously (AND logic), or when any one of the metrics meets its threshold (OR logic). For example, you can trigger a scaling activity only when both CPU utilization and memory usage exceed their thresholds.

2022-07-27

All

• Configure an event-triggered task

• CreateAlarm

• ModifyAlarm

• DescribeAlarms

Scaling configurations support the cloud_auto disk type

You can specify the system disk or data disk type as cloud_auto (ESSD AutoPL disk) in a scaling configuration or launch template to create and attach ECS instances that use this disk type in a scaling group. This disk type maintains the original features and performance of ESSDs while decoupling disk capacity from performance.

2022-07-14

All

• CreateScalingConfiguration

• ModifyScalingConfiguration

• DescribeScalingConfigurations

Feature

Description

Release date

Region

References

ECI instances now support the standby state

You can call the EnterStandby API operation to set an ECI instance in a scaling group to the standby state, or call the ExitStandby API operation to return an ECI instance from the standby state to the running state.

2022-06-29

All

• EnterStandby

• ExitStandby

ECI scaling groups can be associated with ALBs, RDS instances, virtual server groups, and ALB server groups

You can call API operations to attach or detach load balancers, RDS instances, virtual server groups, or ALB server groups from a scaling group for ECI instances.

2022-06-14

All

• AttachLoadBalancers

• DetachLoadBalancers

• AttachDBInstances

• DetachDBInstances

• AttachVServerGroups

• DetachVServerGroups

• AttachAlbServerGroups

• DetachAlbServerGroups

Intelligent configuration supports filter conditions

In intelligent configuration mode, you can set filter conditions to quickly select instance types. For example, you can exclude specific instance types, filter by architecture type, or specify whether to include burstable performance instance types.

2022-06-08

All

• CreateScalingConfiguration

• DescribeScalingConfigurations

• ModifyScalingConfiguration

Agent-based monitoring metrics support a 15-second statistical period

The 15-second statistical period is available only for theCpuUtilizationAgent: (Agent) CPU utilization (%) metric. For all other agent-collected metrics, the available statistical periods are 60, 120, 300, and 900 seconds.

2022-06-08

All

• CreateAlarm

• ModifyAlarm

ScaleWithAdjustment API support for scaling groups with an expected number of instances

If you set a desired capacity when you create a scaling group and have not created any scaling rules, you can call the ScaleWithAdjustment API to quickly execute a scaling activity to automatically scale the scaling group.

2022-06-06

All

ScaleWithAdjustment

Feature

Description

Release date

Region

References

Scaling groups for ECI instances support lifecycle hooks

You can use lifecycle hooks to manage the lifecycle of ECI instances within a scaling group. A lifecycle hook pauses scaling activities automatically triggered by Auto Scaling, placing the affected ECI instances in a pending state. This provides a time window to perform custom actions before the hook times out and the instance lifecycle continues.

2022-05-17

All

• Overview of lifecycle hooks

• Configure a lifecycle hook

Feature

Description

Release date

Region

References

Manually add or remove ECI instances in a scaling group

You can call the AttachInstances API to manually add ECI instances to a scaling group, and call the DetachInstances API to remove existing ECI instances from the scaling group.

2022-04-24

All

• AttachInstances

• DetachInstances

Resource-level and tag-level authorization

You can manage scaling groups by using resource-level and tag-level authorization. This lets you grant different permissions to RAM users to delegate management of your Auto Scaling resources.

2022-04-13

All

• Use resource-based authorization to manage Auto Scaling

• Use tag-based authorization to set management permissions for scaling groups

Feature

Description

Release date

Region

References

Scaling groups for ECI instances support health checks

A scaling group can periodically check the health status of its ECI instances. If an instance is unhealthy, Auto Scaling automatically removes it and launches a replacement to maintain the expected number of instances.

2022-02-24

All

• Configure a scaling group

• Instance lifecycle in a scaling group

Feature

Description

Release date

Region

References

Asynchronous operations

You can now move an ECS instance into or out of the standby state asynchronously, which improves the user experience.

2021-11-29

All

• Manually adjust instance status

• Move an instance out of the standby state

ECS instance resource availability assessment

You can now use the elastic strength feature to assess the resource availability of ECS instances, which increases the success rate of scale-out activities for your scaling group.

2021-11-28

All

Elastic strength

Automated handling for unhealthy scaling groups

The console now notifies you about unhealthy scaling groups so you can quickly identify and resolve issues.

2021-11-16

All

View a scaling group

Capacity reservation parameters

You can now set capacity reservation parameters for ECS instances or ECI instances when you create or modify a scaling configuration.

2021-11-02

All

• Create a scaling configuration (ECS instance)

• Create a scaling configuration (ECI instance)

• Modify a scaling configuration

Feature

Description

Release date

Region

References

Automatic scaling for ECI instances

If you set a scaling group's type to ECI, Auto Scaling automatically scales the ECI instances within it.

2021-10-20

All

• Configure a scaling group

• Create a scaling configuration (ECI instance)

Feature

Description

Release date

Region

References

Intelligent configuration

Scaling configurations now support intelligent configuration, where Auto Scaling automatically selects suitable instance types for scale-out activities.

2021-09-09

All

• Create a scaling configuration (ECS instance)

• Configure a scaling configuration by using specified instance properties

• CreateScalingConfiguration

Feature

Description

Release date

Region

References

Economical mode

The No Fees for Stopped Instances feature is renamed to economical mode.

2021-08-09

All

• Configure a scaling group

• CreateScalingGroup

• RemoveInstances

Feature

Description

Release date

Region

References

ALB server group association

Once associated, ECS instances added to the scaling group are automatically registered as backend servers with the ALB server group to handle distributed requests.

2021-06-11

All

• Configure a scaling group

• AttachAlbServerGroups

Feature

Description

Release date

Region

References

Vertical scaling

You can now adjust the instance type (vCPUs and memory) of pay-as-you-go instances.

2021-03-31

All

Vertical scaling overview

Feature

Description

Release date

Region

References

Weighted instance types in scaling configurations

You can now assign weights to different instance types within a scaling configuration.

2021-02-12

All

• Create a scaling configuration (ECS instance)

• Use performance metrics as a scaling dimension

• CreateScalingConfiguration

Feature

Description

Release date

Region

References

Ordered and unique hostnames

Scaling configurations can now automatically generate ordered and unique hostnames.

2021-01-15

All

• Create a scaling configuration (ECS instance)

• Set rules for ordered and unique instance names or hostnames

• CreateScalingConfiguration

Feature

Description

Release date

Region

References

Scaling configuration

When you use an ESSD as the system disk or data disk, you can now specify its performance level.

2020-11-18

All

• Create a scaling configuration (ECS instance)

• CreateScalingConfiguration

Feature

Description

Release date

Region

References

Event-triggered task

Custom monitoring event-triggered tasks now support additional statistical methods.

2020-10-25

All

• Configure an event-triggered task

• CreateAlarm

Feature

Description

Release date

Region

References

Modify a scaling configuration

You can now modify only the image in a scaling configuration.

2020-09-24

All

Update the image in a scaling configuration

Feature

Description

Release date

Region

References

Rolling upgrade

You can now install software using OOS packages during a rolling upgrade.

2020-08-20

Partial

Rolling upgrade

Expected number of instances

You can now modify the expected number of instances when you remove an instance from a scaling group.

2020-08-12

All

Expected number of instances

Scheduled task

You can now create scheduled tasks that have no end time and a maximum recurrence period of one year.

2020-08-07

All

Configure scheduled tasks

Feature

Description

Release date

Region

References

Modify a scaling group

You can now enable or disable health checks for instances in an existing scaling group.

2020-07-30

All

Modify a scaling group

Feature

Description

Release date

Region

References

Auto Scaling service-linked role

Auto Scaling automatically creates the AliyunServiceRoleForAutoScaling service-linked role when you activate the service.

2020-05-06

All

Service-linked role

Feature

Description

Release date

Region

References

Rolling upgrade

You can now update ECS instance configurations in batches using tasks.

2020-04-26

Partial

Rolling upgrade

Feature

Description

Release date

Region

References

System monitoring event-triggered tasks

System monitoring event-triggered tasks now support more monitoring metrics.

2020-03-17

All

System monitoring event-triggered tasks

Create a scaling group

The scaling group creation process is improved. You can now select existing instances and add tags when you create a scaling group.

2020-03-11

All

Configure a scaling group

Feature

Description

Release date

Region

Related documents

Suspend and resume scaling group processes

You can suspend and resume processes in a scaling group to gain more granular control over scaling activities.

2019-12-03

All regions

• Suspend a scaling group process

• Suspend and resume scaling group processes

Feature

Description

Release date

Region

Related documents

Set the health status of an ECS instance

You can set the health status of an ECS instance in a scaling group via an API call.

2019-10-25

All regions

SetInstanceHealth

Expected number of instances

You can set the expected number of ECS instances in a scaling group to enable flexible and concurrent scaling activities.

2019-10-22

All regions

Expected number of instances

Feature

Description

Release date

Region

Related documents

Create a scaling configuration from a specified instance

When creating a scaling group via an API call, you can specify an instance ID to automatically create a scaling configuration based on that instance.

2019-09-24

All regions

CreateScalingGroup

Feature

Description

Release date

Region

Related documents

Associate a lifecycle hook with an OOS O&M template

You can associate a lifecycle hook with an OOS O&M template. When the hook triggers, it automatically runs the specified template.

2019-08-14

All regions

Configure a lifecycle hook

Feature

Description

Release date

Region

Related documents

Cost optimization policy supports mixed instance allocation

With the cost optimization policy for multi-zone scaling, you can combine pay-as-you-go and spot instances to meet the required capacity.

2019-07-02

All regions

Configure a scaling group

Feature

Description

Release date

Region

Related documents

Step scaling rule

Step scaling rules let you adjust capacity in varying increments based on the size of the metric deviation.

2019-05-20

All regions

Configure a scaling rule

Multiple security groups

You can add automatically created instances to multiple security groups.

2019-05-15

All regions

Create a scaling configuration (ECS instance)

View endpoint information

You can make an API call to list the regions where Auto Scaling is available.

2019-05-05

All regions

DescribeRegions

Feature

Description

Release date

Region

Related documents

Predictive scaling rule

You can create predictive scaling rules to automatically adjust the minimum and maximum number of instances in a scaling group based on historical monitoring data.

2019-04-18

All regions

Scaling rule overview

Scaling configurations support DiskName, DiskDescription, and Encrypted parameters

You can set the name, description, and encryption status for disks in a scaling configuration.

2019-04-04

All regions

Create a scaling configuration (ECS instance)

Feature

Description

Release date

Region

Related documents

Target tracking scaling rule

You can create target tracking scaling rules to automatically scale the number of instances to maintain a metric at a specified target value.

2019-01-11

All regions

Configure a scaling rule

Feature

Description

Release date

Region

Related documents

Modify vSwitches

You can modify the vSwitches that are associated with a scaling group.

2018-12-28

All regions

Modify a scaling group

Cost optimization policy

The multi-zone scaling policy supports the cost optimization policy.

2018-12-16

All regions

• Configure a scaling group

• Set a scaling policy and specify multiple instance types to reduce costs

Specify vServer groups

You can associate a scaling group with specific backend vServer groups from an SLB instance.

2018-12-07

All regions

Attach or detach SLB instances from a scaling group

Feature

Description

Release date

Region

Related documents

Enable No Fees for Stopped Instances for specified instances

You can enable the No Fees for Stopped Instances feature for instances when they are removed from a scaling group.

2018-11-28

All regions

• Configure a scaling group

• CreateScalingGroup

• RemoveInstances

Feature

Description

Release date

Region

Related documents

RDS high-security whitelist

You can add the private IP addresses of ECS instances in a scaling group to an RDS high-security whitelist.

2018-10-19

All regions

Configure a scaling group

Feature

Description

Release date

Region

Related documents

Service availability in the UK (London) region

Auto Scaling is available in the UK (London) region.

2018-09-30

UK (London)

What is Auto Scaling?

Stop instances on scale-in

You can configure a scaling group to stop instances instead of terminating them during scale-in events.

2018-09-13

All regions

Configure a scaling group

Use launch templates as the source of instance configurations

You can use a launch template to define the instance configurations for a scaling group.

2018-09-01

All regions

Configure a scaling group

Feature

Description

Release date

Region

Related documents

Cron expressions for scheduled tasks

You can use cron expressions to define recurring scheduled tasks.

2018-08-20

All regions

Configure scheduled tasks

Expiration notifications for scheduled tasks

You can set event notifications for scheduled task expirations.

2018-08-20

All regions

Configure event notifications

Feature

Description

Release date

Region

Related documents

Modify data disk and spot instance information

You can modify information about data disks and the price limits for spot instances in a scaling configuration.

2018-07-27

All regions

Modify a scaling configuration

Associate RDS instances

You can attach or detach RDS instances from an existing scaling group.

2018-07-11

All regions

Modify a scaling group

Associate SLB instances

You can attach or detach SLB instances from an existing scaling group.

2018-07-04

All regions

Modify a scaling group

Feature

Description

Release date

Region

Related documents

Modify a scaling configuration

You can modify existing scaling configurations.

2018-06-27

All regions

Modify a scaling configuration

Lifecycle hooks

When an ECS instance is automatically launched or terminated, a lifecycle hook can put the instance into a wait state, which allows you to perform custom actions on the instance.

2018-06-01

All regions

Configure a lifecycle hook

Feature

Description

Release date

Region

Related documents

Scaling activity notifications

When a specified type of event occurs, Auto Scaling automatically sends a notification to CloudMonitor or Simple Message Queue (formerly MNS).

2018-04-20

All regions

Message notification overview

Feature

Description

Release date

Region

Related documents

Instance protection

You can protect an instance in a scaling group to prevent it from being terminated during a scale-in.

2018-03-28

All regions

Put an instance into the protected state

Standby state

You can move an instance in a scaling group to the standby state, which allows you to manage its lifecycle independently.

2018-03-15

All regions

Manually adjust the status of an instance

Product

Description

Release date

Region

Related documents

Auto Scaling

Auto Scaling was officially launched.

2015-08-27

Selected regions

What is Auto Scaling?

Product

Description

Release date

Region

Related documents

Auto Scaling

Auto Scaling was released for internal preview.

2014-10-15

Selected regions

What is Auto Scaling?