Auto-Scaling Settings

Scaling Group Management Function

Set the following items to create scaling groups and register them in the stack. You can also configure settings for the health check function, which detects abnormality on scaled-out virtual servers and starts recovery automatically.

Scaling Policy Settings

Specify the following items to set a scaling policy.

Startup Configuration Settings

Define the settings for when a scaling policy is applied and the virtual servers that are added are actually started.

Support of the Alarm Function

The alarm setting of the monitoring service specifies a scaling policy as an action to be taken when the value reaches the threshold. You can adjust auto-scaling according to the workload by setting the thresholds for alarms to call different scaling policies.

Example of Setting Auto-Scaling

An example of a stack definition that describes conditions for auto-scaling is shown below. In this example, the conditions are set as described below.

• The following are defined for the scaling group:

  • Specification of a load balancer in order to distribute the load on the auto-scaled virtual servers (balance the traffic load to port 80 (HTTP))
  • Specification of the maximum number of virtual servers as three
  • Specification of the minimum number of virtual servers as two
  • Specification of the subnet to which auto-scaled virtual servers are connected
  • Specification of the startup configuration (specification of values by using variables that are declared in the parameters section)

• The following policies are defined as the scaling policies:

  • web_server_scaleout_policy: Specification of "ChangeInCapacity" for the scaling type, and setting of one (+1) for the number of virtual servers to be added when the alarm is raised
  • web_server_scalein_policy: Specification of "ChangeInCapacity" for the scaling type, and setting of one (-1) for the number of virtual servers to be deleted when the alarm is raised

• The following two alarms are defined as the alarms:

  • cpu_alarm_high: Application of web_server_scaleout_policy when a CPU usage rate of higher than 50% that continues for one minute or more is detected
  • cpu_alarm_low: Application of web_server_scalein_policy when a CPU usage rate of 15% or lower that continues for one minute or more is detected

Example of stack definition:

heat_template_version: 2013-05-23 

description: 
  Autoscaling sample template. 

parameters: 

  az: 
    type: string 
    default: jp-east-1a 
  param_image_id: 
    type: string 
# ImageID of CentOS 
    default: 1234abcd-5678-ef90-9876-fedc5432dcba 
  param_flavor: 
    type: string 
    default: standard 

  key_name: 
    type: string 
    description: SSH key to connect to the servers 
    default: sample_keypair00 
  autoscale_security_group: 
    type: comma_delimited_list 
    default: sample_SG00 


resources: 

  web_server_group: 
    type: FCX::AutoScaling::AutoScalingGroup 
    properties: 
      AvailabilityZones: [{get_param: az}] 
      LaunchConfigurationName: {get_resource: launch_config} 
      MinSize: '2' 
      MaxSize: '3' 
# subnet ID for auto-scaling 
      VPCZoneIdentifier: [38e6630f-3257-4ee8-a006-f6d57ceaa2c3] 
      LoadBalancerNames: 
      - {get_resource: fj_elb} 

  launch_config: 
    type: FCX::AutoScaling::LaunchConfiguration 
    properties: 
      ImageId: { get_param: param_image_id } 
      InstanceType: { get_param: param_flavor } 
      KeyName: {get_param: key_name} 
      SecurityGroups: {get_param: autoscale_security_group} 
      BlockDeviceMappingsV2: [{source_type: 'image', destination_type: 'volume', boot_index: '0', device_name: '/dev/vda',  volume_size: '40', uuid: {get_param: param_image_id}, delete_on_termination: true}] 

  fj_elb: 
    type: FCX::ExpandableLoadBalancer::LoadBalancer 
    properties: 
# subnet ID for auto-scaling 
      Subnets: [38e6630f-3257-4ee8-a006-f6d57ceaa2c3] 
      Listeners: 
      - {LoadBalancerPort: '80', InstancePort: '80', 
        Protocol: 'HTTP', InstanceProtocol: 'HTTP' } 
      HealthCheck: {Target: 'HTTP:80/healthcheck', HealthyThreshold: '3', 
                   UnhealthyThreshold: '5', Interval: '30', Timeout: '5'} 
      Version: 2014-09-30 
      Scheme: internal 
      LoadBalancerName: fjsampleELBaz1 

  web_server_scaleout_policy: 
    type: FCX::AutoScaling::ScalingPolicy 
    properties: 
      AdjustmentType: ChangeInCapacity 
      AutoScalingGroupName: {get_resource: web_server_group} 
      Cooldown: '60' 
      ScalingAdjustment: '1' 

  web_server_scalein_policy: 
    type: FCX::AutoScaling::ScalingPolicy 
    properties: 
      AdjustmentType: ChangeInCapacity 
      AutoScalingGroupName: {get_resource: web_server_group} 
      Cooldown: '60' 
      ScalingAdjustment: '-1' 

  cpu_alarm_high: 
    type: OS::Ceilometer::Alarm 
    properties: 
      description: Scale-out if the average CPU > 50% for 1 minute 
      meter_name: fcx.compute.cpu_util 
      statistic: avg 
      period: '60' 
      evaluation_periods: '1' 
      threshold: '50' 
      alarm_actions: 
      - {get_attr: [web_server_scaleout_policy, AlarmUrl]} 
      matching_metadata: {'metadata.user_metadata.groupname': {get_resource: 'web_server_group'}} 
      comparison_operator: gt 

  cpu_alarm_low: 
    type: OS::Ceilometer::Alarm 
    properties: 
      description: Scale-in if the average CPU < 15% for 1 minute 
      meter_name: fcx.compute.cpu_util 
      statistic: avg 
      period: '60' 
      evaluation_periods: '1' 
      threshold: '15' 
      alarm_actions: 
      - {get_attr: [web_server_scalein_policy, AlarmUrl]} 
      matching_metadata: {'metadata.user_metadata.groupname': {get_resource: 'web_server_group'}} 
      comparison_operator: lt