|« PreviousNext »|
|Did this page help you? Yes | No | Tell us about it...|
Auto Scaling is an Amazon web service designed to automatically launch or terminate Amazon EC2 instances based on user-defined triggers. Users can set up Auto Scaling groups and associate triggers with these groups to automatically scale computing resources based on metrics such as bandwidth usage or CPU utilization. Auto Scaling works with Amazon CloudWatch to retrieve metrics for the server instances running your application.
Auto Scaling lets you take a group of Amazon EC2 instances and set various parameters to have this group automatically increase or decrease in number. Auto Scaling can add or remove Amazon EC2 instances from that group to help you seamlessly deal with traffic changes to your application.
Auto Scaling also monitors the health of each Amazon EC2 instance that it launches. If any instance terminates unexpectedly, Auto Scaling detects the termination and launches a replacement instance. This capability enables you to maintain a fixed, desired number of Amazon EC2 instances automatically.
AWS Elastic Beanstalk provisions Auto Scaling for your application. You can edit the AWS Elastic Beanstalk environment's Amazon EC2 instance configuration with the Auto Scaling tab inside your application environment tab in the AWS Toolkit for Visual Studio.
The following section discusses how to configure Auto Scaling parameters for your application.
You can edit the launch configuration to control how your AWS Elastic Beanstalk application provisions Auto Scaling resources.
The Minimum Instance Count and Maximum Instance Count text boxes let you specify the minimum and maximum size of the Auto Scaling group that your AWS Elastic Beanstalk application uses.
To maintain a fixed number of Amazon EC2 instances, set the Minimum Instance Count and Maximum Instance Count text boxes to the same value.
The Availability Zones text box lets you specify the number of Availability Zones you want your Amazon EC2 instances to be in. It is important to set this number if you want to build fault-tolerant applications. If one Availability Zone goes down, your instances will still be running in your other Availability Zones.
Currently, it is not possible to specify which Availability Zone your instance will be in.
A trigger is an Auto Scaling mechanism that you set to tell the system when you want to increase (scale out) the number of instances, and when you want to decrease (scale in) the number of instances. You can configure triggers to fire on any metric published to Amazon CloudWatch, such as CPU utilization, and determine if the conditions you specified have been met. When the upper or lower thresholds of the conditions you have specified for the metric have been breached for the specified period of time, the trigger launches a long-running process called a Scaling Activity.
You can define a scaling trigger for your AWS Elastic Beanstalk application using AWS Toolkit for Visual Studio.
Auto Scaling triggers work by watching a specific Amazon CloudWatch metric for an instance. Triggers include CPU utilization, network traffic, and disk activity. Use the Trigger Measurement drop-down list box to select a metric for your trigger.
The following list describes the trigger parameters you can configure using the AWS Management Console.
You can specify which statistic the trigger should use. You can select
the Trigger Statistic drop-down list.
Specify the unit for the trigger measurement using the Unit of Measurement drop-down list.
The value in the Measurement Period text box specifies how frequently Amazon CloudWatch measures the metrics for your trigger. The Breach Duration is the amount of time a metric can be beyond its defined limit (as specified in the Upper Threshold and Lower Threshold text boxes) before the trigger fires.
The Upper Breach Scale Increment and Lower Breach Scale Increment text boxes specify how many Amazon EC2 instances to add or remove when performing a scaling activity.
For more information on Auto Scaling, go to the Auto Scaling documentation.