# Understand automatic scaling for Spot Fleet
**Important**
Spot Fleet uses a legacy API with no planned investment. We recommend using EC2 Fleet or an Auto Scaling group instead. For more information, see [Which is the best fleet method to use?](which-fleet-method-to-use.md).
*Automatic scaling* enables your Spot Fleet to increase or decrease its target capacity based on demand. With automatic scaling, a Spot Fleet can either launch instances (scale out) or terminate instances (scale in) within a specified range, in response to one or more scaling policies.
Automatic scaling for Spot Fleet is made possible by a combination of the Amazon EC2, Amazon CloudWatch, and Application Auto Scaling APIs. Spot Fleet requests are created with Amazon EC2, alarms are created with CloudWatch, and scaling policies are created with Application Auto Scaling.
**Types of automatic scaling**
Spot Fleet supports the following types of automatic scaling:
+ [Target tracking scaling](spot-fleet-target-tracking.md) – Increase or decrease
the current capacity of the fleet by targeting a value for a specific metric. This is similar to the way that your thermostat maintains the temperature of your home—you select the desired temperature and the thermostat does the rest.
+ [Step scaling](spot-fleet-step-scaling.md) – Increase or decrease the current capacity of the fleet based on a set of scaling adjustments, known as step adjustments, that vary based on the size of the alarm breach.
+ [Scheduled scaling](spot-fleet-scheduled-scaling.md) – Increase or decrease the current capacity of the fleet based on the date and time.
## Considerations
When using automatic scaling for your Spot Fleet, consider the following:
+ **Instance weighting** – If you're using [instance weighting](ec2-fleet-instance-weighting.md), keep in mind that Spot Fleet can exceed the target capacity as needed. Fulfilled capacity can be a floating-point number but target capacity must be an integer, so Spot Fleet rounds up to the next integer. You must take these behaviors into account when you look at the outcome of a scaling policy when an alarm is triggered. For example, suppose that the target capacity is 30, the fulfilled capacity is 30.1, and the scaling policy subtracts 1. When the alarm is triggered, the automatic scaling process subtracts 1 from 30.1 to get 29.1 and then rounds it up to 30, so no scaling action is taken. As another example, suppose that you selected instance weights of 2, 4, and 8, and a target capacity of 10, but no weight 2 instances were available so Spot Fleet provisioned instances of weights 4 and 8 for a fulfilled capacity of 12. If the scaling policy decreases target capacity by 20% and an alarm is triggered, the automatic scaling process subtracts 12\$10.2 from 12 to get 9.6 and then rounds it up to 10, so no scaling action is taken.
+ **Cooldown period** – The scaling policies that you create for Spot Fleet support a cooldown period. This is the number of seconds after a scaling activity completes where previous trigger-related scaling activities can influence future scaling events. For scale-out policies, while the cooldown period is in effect, the capacity that has been added by the previous scale-out event that initiated the cooldown is calculated as part of the desired capacity for the next scale out. The intention is to continuously (but not excessively) scale out. For scale in policies, the cooldown period is used to block subsequent scale in requests until it has expired. The intention is to scale in conservatively to protect your application's availability. However, if another alarm triggers a scale-out policy during the cooldown period after a scale-in, automatic scaling scales out your scalable target immediately.
+ **Use detailed monitoring** – We recommend that you scale based on instance metrics with a 1-minute frequency because that ensures a faster response to utilization changes. Scaling on metrics with a 5-minute frequency can result in slower response time and scaling on stale metric data. To send metric data for your instances to CloudWatch in 1-minute periods, you must specifically enable detailed monitoring. For more information, see [Manage detailed monitoring for your EC2 instances](manage-detailed-monitoring.md) and [Create a Spot Fleet request using defined parameters](create-spot-fleet.md#create-spot-fleet-advanced).
+ **AWS CLI** – If you use the AWS CLI for configuring scaling for Spot Fleet, you'll use the [application-autoscaling](https://docs.aws.amazon.com/cli/latest/reference/application-autoscaling) commands.
# IAM permissions required for Spot Fleet automatic scaling
Automatic scaling for Spot Fleet is made possible by a combination of the Amazon EC2, Amazon CloudWatch, and Application Auto Scaling APIs. Spot Fleet requests are created with Amazon EC2, alarms are created with CloudWatch, and scaling policies are created with Application Auto Scaling. In addition to the [IAM permissions required for using Spot Fleet](spot-fleet-prerequisites.md#spot-fleet-iam-users) and Amazon EC2, the user that accesses the fleet scaling settings must have the appropriate permissions for the services that support automatic scaling.
To use automatic scaling for Spot Fleet, users must have permissions to use the actions shown in the following example policy.
------
#### [ JSON ]
****
```
{
"Version":"2012-10-17",
"Statement": [
{
"Effect": "Allow",
"Action": [
"application-autoscaling:*",
"ec2:DescribeSpotFleetRequests",
"ec2:ModifySpotFleetRequest",
"cloudwatch:DeleteAlarms",
"cloudwatch:DescribeAlarmHistory",
"cloudwatch:DescribeAlarms",
"cloudwatch:DescribeAlarmsForMetric",
"cloudwatch:GetMetricStatistics",
"cloudwatch:ListMetrics",
"cloudwatch:PutMetricAlarm",
"cloudwatch:DisableAlarmActions",
"cloudwatch:EnableAlarmActions",
"iam:CreateServiceLinkedRole",
"sns:CreateTopic",
"sns:Subscribe",
"sns:Get*",
"sns:List*"
],
"Resource": "*"
}
]
}
```
------
You can also create your own IAM policies that allow more fine-grained permissions for calls to the Application Auto Scaling API. For more information, see [Identity and Access Management for Application Auto Scaling](https://docs.aws.amazon.com/autoscaling/application/userguide/auth-and-access-control.html) in the *Application Auto Scaling User Guide*.
The Application Auto Scaling service also needs permission to describe your Spot Fleet and CloudWatch alarms, and permissions to modify your Spot Fleet target capacity on your behalf. If you enable automatic scaling for your Spot Fleet, it creates a service-linked role named `AWSServiceRoleForApplicationAutoScaling_EC2SpotFleetRequest`. This service-linked role grants Application Auto Scaling permission to describe the alarms for your policies, to monitor the current capacity of the fleet, and to modify the capacity of the fleet. The original managed Spot Fleet role for Application Auto Scaling was `aws-ec2-spot-fleet-autoscale-role`, but it is no longer required. The service-linked role is the default role for Application Auto Scaling. For more information, see [Service-linked roles for Application Auto Scaling](https://docs.aws.amazon.com/autoscaling/application/userguide/application-auto-scaling-service-linked-roles.html) in the *Application Auto Scaling User Guide*.
# Target tracking scaling: Scale Spot Fleet by targeting a value for a specific metric
With target tracking scaling, you create a target tracking scaling policy by selecting a metric and setting a target value. Spot Fleet then creates and manages the CloudWatch alarms that trigger the scaling policy, and calculates the scaling adjustment based on the chosen metric and target value. The scaling policy adjusts capacity by adding or removing instances as needed to keep the metric at, or close to, the specified target value. A target tracking policy not only keeps the metric close to the target value, but also adjusts to the fluctuations in the metric due to a fluctuating load pattern and minimizes rapid capacity fluctuations.
You can create multiple target tracking scaling policies for a Spot Fleet, provided each policy uses a different metric. The fleet scales based on the policy that specifies the largest fleet capacity. This allows you to cover multiple scenarios to ensure sufficient capacity for your application workloads.
To ensure application availability, the fleet scales out proportionally to the metric as fast as it can, but scales in more gradually.
When a Spot Fleet terminates a Spot Instance because the target capacity was decreased, the instance receives a Spot Instance interruption notice.
**Note**
Do not edit or delete the CloudWatch alarms that Spot Fleet manages for a target tracking scaling policy. Spot Fleet deletes the alarms automatically when you delete the target tracking scaling policy.
**Prerequisites**
+ The Spot Fleet request must have a request type of `maintain`. Automatic scaling is not supported for requests of type `request`.
+ Configure the [IAM permissions required for Spot Fleet automatic scaling](spot-fleet-auto-scaling-IAM.md).
+ Review the [Considerations](spot-fleet-automatic-scaling.md#considerations-for-spot-fleet-automatic-scaling).
**To configure a target tracking policy**
1. Open the Amazon EC2 console at [https://console.aws.amazon.com/ec2/](https://console.aws.amazon.com/ec2/).
1. In the navigation pane, choose **Spot Requests**.
1. Select your Spot Fleet request.
1. Choose the **Auto Scaling** tab near the bottom of the screen. If you selected the link for your Spot Fleet, there is no tab; instead, scroll down to the **Auto Scaling** section.
1. If automatic scaling is not configured, choose **Configure**.
1. Use **Scale capacity between** to set the minimum and maximum capacity for your fleet. Automatic scaling does not scale your fleet below the minimum capacity or above the maximum capacity.
1. For **Policy name**, enter a name for the policy.
1. Choose a **Target metric**.
1. Enter a **Target value** for the metric.
1. For **Cooldown period**, specify a new value (in seconds) or keep the default.
1. (Optional) To omit creating a scale-in policy based on the current configuration, select **Disable scale-in**. You can create a scale-in policy using a different configuration.
1. Choose **Save**.
**To configure a target tracking policy using the AWS CLI**
1. Register the Spot Fleet request as a scalable target using the [register-scalable-target](https://docs.aws.amazon.com/cli/latest/reference/application-autoscaling/register-scalable-target.html) command.
1. Create a scaling policy using the [put-scaling-policy](https://docs.aws.amazon.com/cli/latest/reference/application-autoscaling/put-scaling-policy.html) command.
# Step scaling: Scale Spot Fleet using step scaling policies
With step scaling policies, you specify CloudWatch alarms to trigger the scaling process. For example, if you want to scale out when CPU utilization reaches a certain level, create an alarm using the `CPUUtilization` metric provided by Amazon EC2.
When you create a step scaling policy, you must specify one of the following scaling adjustment types:
+ **Add** – Increase the target capacity of the fleet by a specified number of capacity units or a specified percentage of the current capacity.
+ **Remove** – Decrease the target capacity of the fleet by a specified number of capacity units or a specified percentage of the current capacity.
+ **Set to** – Set the target capacity of the fleet to the specified number of capacity units.
When an alarm is triggered, the automatic scaling process calculates the new target capacity using the fulfilled capacity and the scaling policy, and then updates the target capacity accordingly. For example, suppose that the target capacity and fulfilled capacity are 10 and the scaling policy adds 1. When the alarm is triggered, the automatic scaling process adds 1 to 10 to get 11, so Spot Fleet launches 1 instance.
When a Spot Fleet terminates a Spot Instance because the target capacity was decreased, the instance receives a Spot Instance interruption notice.
**Prerequisites**
+ The Spot Fleet request must have a request type of `maintain`. Automatic scaling is not supported for requests of type `request`.
+ Configure the [IAM permissions required for Spot Fleet automatic scaling](spot-fleet-auto-scaling-IAM.md).
+ Consider which CloudWatch metrics are important to your application. You can create CloudWatch alarms based on metrics provided by AWS or your own custom metrics.
+ For the AWS metrics that you will use in your scaling policies, enable CloudWatch metrics collection if the service that provides the metrics does not enable it by default.
+ Review the [Considerations](spot-fleet-automatic-scaling.md#considerations-for-spot-fleet-automatic-scaling).
**To create a CloudWatch alarm**
1. Open the CloudWatch console at [https://console.aws.amazon.com/cloudwatch/](https://console.aws.amazon.com/cloudwatch/).
1. In the navigation pane, expand **Alarms** and choose **All alarms**.
1. Choose **Create alarm**.
1. On the **Specify metric and conditions** page, choose **Select metric**.
1. Choose **EC2 Spot**, then **Fleet Request Metrics**, and then select a metric (for example, **TargetCapacity**), and then choose **Select metric**.
The **Specify metric and conditions** page appears, showing a graph and other information about the metric you selected.
1. For **Period**, choose the evaluation period for the alarm, for example, **1 minute**. When evaluating the alarm, each period is aggregated into one data point.
**Note**
A shorter period creates a more sensitive alarm.
1. For **Conditions**, define the alarm by defining the threshold condition. For example, you can define a threshold to trigger the alarm whenever the value of the metric is greater than or equal to 80 percent.
1. Under **Additional configuration**, for **Datapoints to alarm**, specify how many datapoints (evaluation periods) must be in the ALARM state to trigger the alarm, for example, 1 evaluation period or 2 out of 3 evaluation periods. This creates an alarm that goes to ALARM state if that many consecutive periods are breaching. For more information, see [Evaluating an alarm](https://docs.aws.amazon.com/AmazonCloudWatch/latest/monitoring/AlarmThatSendsEmail.html#alarm-evaluation) in the *Amazon CloudWatch User Guide*.
1. For **Missing data treatment**, choose one of the options (or leave the default of **Treat missing data as missing**). For more information, see [Configuring how CloudWatch alarms treat missing data](https://docs.aws.amazon.com/AmazonCloudWatch/latest/monitoring/AlarmThatSendsEmail.html#alarms-and-missing-data) in the *Amazon CloudWatch User Guide*.
1. Choose **Next**.
1. (Optional) To receive notification of a scaling event, for **Notification**, you can choose or create the Amazon SNS topic you want to use to receive notifications. Otherwise, you can delete the notification now and add one later as needed.
1. Choose **Next**.
1. Under **Add name and description**, enter a name and description for the alarm and choose **Next**.
1. Choose **Create alarm**.
**To configure a step scaling policy for your Spot Fleet**
1. Open the Amazon EC2 console at [https://console.aws.amazon.com/ec2/](https://console.aws.amazon.com/ec2/).
1. In the navigation pane, choose **Spot Requests**.
1. Select your Spot Fleet request.
1. Choose the **Auto Scaling** tab near the bottom of the screen. If you selected the link for your Spot Fleet, there is no tab; instead, scroll down to the **Auto Scaling** section.
1. If automatic scaling is not configured, choose **Configure**.
1. Use **Scale capacity between** to set the minimum and maximum capacity for your fleet. Scaling policies do not scale your fleet below the minimum capacity or above the maximum capacity.
1. Under **Scaling policies**, for **Policy type**, choose **Step scaling policy**.
1. Initially, **Scaling policies** contain step scaling policies named **ScaleUp** and **ScaleDown**. You can complete these policies, or choose **Remove policy** to delete them. You can also choose **Add policy**.
1. To define a policy, do the following:
1. For **Policy name**, enter a name for the policy.
1. For **Policy trigger**, select an existing alarm, or choose **Create alarm** to open the Amazon CloudWatch console and create an alarm.
1. For **Modify capacity**, define the amount by which to scale and the lower and upper bound of the step adjustment. You can add or remove a specific number of instances or a percentage of the existing fleet size, or set the fleet to an exact size.
For example, to create a step scaling policy that increases the capacity of the fleet by 30 percent, choose **Add**, enter **30** in the next field, and then choose **percent**. By default, the lower bound for an add policy is the alarm threshold and the upper bound is positive (\$1) infinity. By default, the upper bound for a remove policy is the alarm threshold and the lower bound is negative (-) infinity.
1. (Optional) To add another step, choose **Add step**.
1. For **Cooldown period**, specify a new value (in seconds) or keep the default.
1. Choose **Save**.
**To configure step scaling policies for your Spot Fleet using the AWS CLI**
1. Register the Spot Fleet request as a scalable target using the [register-scalable-target](https://docs.aws.amazon.com/cli/latest/reference/application-autoscaling/register-scalable-target.html) command.
1. Create a scaling policy using the [put-scaling-policy](https://docs.aws.amazon.com/cli/latest/reference/application-autoscaling/put-scaling-policy.html) command.
1. Create an alarm that triggers the scaling policy using the [put-metric-alarm](https://docs.aws.amazon.com/cli/latest/reference/cloudwatch/put-metric-alarm.html) command.
# Scheduled scaling: Scale Spot Fleet on a schedule
Scaling your fleet on a schedule enables you to scale your application in response to predictable changes in demand. By creating *scheduled actions*, you can instruct Spot Fleet to perform scaling activities at specific times. To create a scheduled action, you must specify an existing Spot Fleet, the time when the scaling activity must occur, and the desired minimum and maximum capacity. Scheduled actions can be configured to scale once or on a recurring schedule. If you needs change, you can edit or delete scheduled actions.
**Prerequisites**
+ Scheduled actions can only be created for existing Spot Fleets. You can't create a scheduled action when you create a Spot Fleet.
+ The Spot Fleet request must have a request type of `maintain`. Automatic scaling is not supported for requests of type `request`.
+ Configure the [IAM permissions required for Spot Fleet automatic scaling](spot-fleet-auto-scaling-IAM.md).
+ Review the [Considerations](spot-fleet-automatic-scaling.md#considerations-for-spot-fleet-automatic-scaling).
**To create a one-time scheduled action**
1. Open the Amazon EC2 console at [https://console.aws.amazon.com/ec2/](https://console.aws.amazon.com/ec2/).
1. In the navigation pane, choose **Spot Requests**.
1. Select your Spot Fleet request.
1. Choose the **Scheduled Scaling** tab near the bottom of the screen. If you selected the link for your Spot Fleet, there is no tab; instead, scroll down to the **Scheduled Scaling** section.
1. Choose **Create scheduled action**.
1. For **Name**, specify a name for the scheduled action.
1. Enter a value for **Minimum capacity**, **Maximum capacity**, or both.
1. For **Recurrence**, choose **Once**.
1. (Optional) Choose a date and time for **Start time**, **End time**, or both.
1. Choose **Create**.
**To create a recurring scheduled action**
1. Open the Amazon EC2 console at [https://console.aws.amazon.com/ec2/](https://console.aws.amazon.com/ec2/).
1. In the navigation pane, choose **Spot Requests**.
1. Select your Spot Fleet request.
1. Choose the **Scheduled Scaling** tab near the bottom of the screen. If you selected the link for your Spot Fleet, there is no tab; instead, scroll down to the **Scheduled Scaling** section.
1. For **Name**, specify a name for the scheduled action.
1. Enter a value for **Minimum capacity**, **Maximum capacity**, or both.
1. For **Recurrence**, choose one of the predefined schedules (for example, **Every day**), or choose **Custom** and enter a cron expression. For more information about the cron expressions supported by scheduled scaling, see [Cron expressions](https://docs.aws.amazon.com/eventbridge/latest/userguide/eb-scheduled-rule-pattern.html#eb-cron-expressions) in the *Amazon EventBridge User Guide*.
1. (Optional) Choose a date and time for **Start time**, **End time**, or both.
1. Choose **Submit**.
**To edit a scheduled action**
1. Open the Amazon EC2 console at [https://console.aws.amazon.com/ec2/](https://console.aws.amazon.com/ec2/).
1. In the navigation pane, choose **Spot Requests**.
1. Select your Spot Fleet request.
1. Choose the **Scheduled Scaling** tab near the bottom of the screen. If you selected the link for your Spot Fleet, there is no tab; instead, scroll down to the **Scheduled Scaling** section.
1. Select the scheduled action and choose **Actions**, **Edit**.
1. Make the needed changes and choose **Submit**.
**To delete a scheduled action**
1. Open the Amazon EC2 console at [https://console.aws.amazon.com/ec2/](https://console.aws.amazon.com/ec2/).
1. In the navigation pane, choose **Spot Requests**.
1. Select your Spot Fleet request.
1. Choose the **Scheduled Scaling** tab near the bottom of the screen. If you selected the link for your Spot Fleet, there is no tab; instead, scroll down to the **Scheduled Scaling** section.
1. Select the scheduled action and choose **Actions**, **Delete**.
1. When prompted for confirmation, choose **Delete**.
**To manage scheduled scaling using the AWS CLI**
Use the following commands:
+ [put-scheduled-action](https://docs.aws.amazon.com/cli/latest/reference/application-autoscaling/put-scheduled-action.html)
+ [describe-scheduled-actions](https://docs.aws.amazon.com/cli/latest/reference/application-autoscaling/describe-scheduled-actions.html)
+ [delete-scheduled-action](https://docs.aws.amazon.com/cli/latest/reference/application-autoscaling/delete-scheduled-action.html)