# Create a launch template using advanced settings Create a launch template using advanced settings This topic describes how to create a launch template with advanced settings from the AWS Management Console. **To create a launch template using advanced settings** 1. Open the Amazon EC2 console at [https://console.aws.amazon.com/ec2/](https://console.aws.amazon.com/ec2/). 1. On the navigation pane, under **Instances**, choose **Launch Templates**, and then choose **Create launch template**. 1. Configure your launch template as described in the following topics: + [Required settings](#configure-required-settings) + [Advanced settings](#configure-advanced-settings) 1. Choose **Create launch template**. ## Required settings When you create a launch template, you must include the following required settings. **Launch template name** Enter a unique name that describes the launch template. **Application and OS Images (Amazon Machine Image)** Choose the Amazon Machine Image (AMI) that you want to use. You can either search or browse for the AMI you want to use. For best scaling efficiency, choose a custom AMI that is fully configured to launch an instance with your application code and requires few modifications on launch. **Instance type** Choose an instance type that is compatible with your AMI. You can skip adding an instance type to your launch template if you plan to use multiple instances types that are embedded in the Auto Scaling group's own resource definition. An instance type is only required if you don't plan to create a [mixed instances group](ec2-auto-scaling-mixed-instances-groups.md). ## Advanced settings The advanced settings are optional. If you do not configure any advanced settings, the specific capabilities will not be added to your instances. Expand the **Advanced details** section to view the advanced settings. The following sections describe the most useful advanced settings to focus on when creating a launch template for an Auto Scaling group. For more information, see [Advanced details](https://docs.aws.amazon.com/AWSEC2/latest/UserGuide/create-launch-template.html#lt-advanced-details) in the *Amazon EC2 User Guide*. **IAM instance profile** The instance profile contains the IAM role that you want to use. When your Auto Scaling group launches an EC2 instance, the permissions defined in the associated IAM role are granted to applications running on the instance. For more information, see [IAM role for applications that run on Amazon EC2 instances](us-iam-role.md). **Termination protection** When enabled, this feature prevents users from terminating an instance using the Amazon EC2 console, CLI commands, and API operations. Termination protection provides an extra safeguard against accidental termination. It does not prevent Amazon EC2 Auto Scaling from terminating an instance. To control which instances Amazon EC2 Auto Scaling can terminate, see [Use instance scale-in protection to control instance termination](ec2-auto-scaling-instance-protection.md). **Detailed CloudWatch monitoring** You can enable detailed monitoring for your EC2 instances to allow them to send metric data to Amazon CloudWatch at 1-minute intervals. By default, EC2 instances send metric data to CloudWatch at 5-minute intervals. Additional charges apply. For more information, see [Configure monitoring for Auto Scaling instances](enable-as-instance-metrics.md). **Credit specification** Amazon EC2 provides burstable performance instances, such as T2, T3, and T3a, that allow applications to burst beyond the baseline CPU performance when required. By default, these instances can burst for a limited time before their CPU usage is throttled. You can optionally enable unlimited mode so that the instances can burst beyond the baseline for as long as needed. This allows applications to sustain high CPU performance when required. Additional charges may apply. For more information, see [Use an Auto Scaling group to launch a burstable performance instance as Unlimited](https://docs.aws.amazon.com/AWSEC2/latest/UserGuide/burstable-performance-instances-how-to.html#burstable-performance-instances-auto-scaling-grp) in the *Amazon EC2 User Guide*. **Placement group name** You can specify a placement group and use a cluster or a partition strategy to influence how your instances are physically located in the AWS data center. For small Auto Scaling groups, you can also use the spread strategy. For more information, see [Placement groups](https://docs.aws.amazon.com/AWSEC2/latest/UserGuide/placement-groups.html) in the *Amazon EC2 User Guide*. There are some considerations when using placement groups with Auto Scaling groups: + If a placement group is specified in both the launch template and the Auto Scaling group, the placement group for the Auto Scaling group takes precedence. + In CloudFormation, be careful if you define a placement group in the launch template. Amazon EC2 Auto Scaling will launch instances into the specified placement group. However, CloudFormation will not receive signals from those instances if you use an [UpdatePolicy](https://docs.aws.amazon.com/AWSCloudFormation/latest/UserGuide/aws-attribute-updatepolicy.html) with your Auto Scaling group (though this could change in the future). **Purchasing option** You can choose **Request Spot Instances** to request Spot Instances at the Spot price, capped at the On-Demand price, and choose **Customize** to change the default Spot Instance settings. For an Auto Scaling group, you must specify a one-time request with no end date (the default). For more information, see [Request Spot Instances for fault-tolerant and flexible applications](launch-template-spot-instances.md). This setting may be useful in special circumstances, but in general it's best to leave it unspecified and create a mixed instances group instead. For more information, see [Auto Scaling groups with multiple instance types and purchase options](ec2-auto-scaling-mixed-instances-groups.md). If you specify a Spot Instance request in your launch template, you can't create a mixed instances group. If you try to use a launch template that requests Spot Instances with a mixed instances group, you receive the following error message: `Incompatible launch template: You cannot use a launch template that is set to request Spot Instances (InstanceMarketOptions) when you configure an Auto Scaling group with a mixed instances policy. Add a different launch template to the group and try again.` **Capacity Reservation** Capacity Reservations allow you to reserve capacity for your Amazon EC2 instances in a specific Availability Zone for any duration. For more information, see [On-Demand Capacity Reservations](https://docs.aws.amazon.com/AWSEC2/latest/UserGuide/ec2-capacity-reservations.html) in the *Amazon EC2 User Guide*. You can choose whether to launch instances into: + any open Capacity Reservation (**Open**) + a specific Capacity Reservation (**Target by ID**) + a group of Capacity Reservations (**Target by group**) To target a specific Capacity Reservation, the instance type in your launch template must match the instance type of the reservation. When you create your Auto Scaling group, use the same Availability Zone as the Capacity Reservation. Depending on the AWS Region you choose, you can choose to target a Capacity Block instead. For more information, see [Use Capacity Blocks for machine learning workloads](launch-template-capacity-blocks.md). To target a group of Capacity Reservations, see [Reserve capacity in specific Availability Zones with Capacity Reservations](use-ec2-capacity-reservations.md). By targeting a group of Capacity Reservations, you can have capacity distributed across multiple Availability Zones to improve resiliency. **Tenancy** Amazon EC2 provides three options for the tenancy of your EC2 instances: + Shared (**Shared**) – Multiple AWS accounts may share the same physical hardware. This is the default tenancy option when launching an instance. + Dedicated instances (**Dedicated**) – Your instance runs on single-tenant hardware. No other AWS customer shares the same physical server. For more information, see [Dedicated Instances](https://docs.aws.amazon.com/AWSEC2/latest/UserGuide/dedicated-instance.html) in the *Amazon EC2 User Guide*. + Dedicated Hosts (**Dedicated host**) – The instance runs on a physical server that is dedicated to your use. Using Dedicated Hosts makes it easier to bring your own licenses (BYOL) that have dedicated hardware requirements to EC2 and meet compliance use cases. If you choose this option, you must provide a host resource group for **Tenancy host resource group**. For more information, see [Dedicated Hosts](https://docs.aws.amazon.com/AWSEC2/latest/UserGuide/dedicated-hosts-overview.html) in the *Amazon EC2 User Guide*. Support for Dedicated Hosts is only available if you specify a host resource group. You can't target a specific host ID or use host placement affinity. + If you try to use a launch template that specifies a host ID, you receive the following error message: `Incompatible launch template: Tenancy host ID is not supported for Auto Scaling.` + If you try to use a launch template that specifies host placement affinity, you receive the following error message: `Incompatible launch template: Auto Scaling does not support host placement affinity.` **Tenancy host resource group** With AWS License Manager, you can bring your own licenses to AWS and manage them centrally. A host resource group is a group of Dedicated Hosts that are linked to a specific License Manager license configuration. Host resource groups allow you to easily launch EC2 instances onto Dedicated Hosts that match your software licensing needs. You do not need to manually allocate Dedicated Hosts ahead of time. They are automatically created as needed. Note that when you associate an AMI with a license configuration, that AMI can only be associated with one host resource group at a time. For more information, see [Host resource groups in AWS License Manager](https://docs.aws.amazon.com/license-manager/latest/userguide/host-resource-groups.html) in the *License Manager User Guide*. **License configurations** With this setting, you can specify a license configuration for your instances without restricting their tenancy to Dedicated Hosts. The license configuration tracks the software licenses deployed on the instances so you can monitor your license usage and compliance. For more information, see [Create a self-managed license](https://docs.aws.amazon.com/license-manager/latest/userguide/create-license-configuration.html) in the *License Manager User Guide*. **Metadata accessible** You can choose whether to enable or disable access to the HTTP endpoint of the instance metadata service. By default, the HTTP endpoint is enabled. If you choose to disable the endpoint, access to your instance metadata is turned off. You can specify the condition to require IMDSv2 only when the HTTP endpoint is enabled. For more information, see [Configure the instance metadata options](https://docs.aws.amazon.com/AWSEC2/latest/UserGuide/configuring-instance-metadata-options.html) in the *Amazon EC2 User Guide*. **Metadata version** You can choose to require the use of Instance Metadata Service Version 2 (IMDSv2) when requesting instance metadata. If you do not specify a value, the default is to support both IMDSv1 and IMDSv2. For more information, see [Configure the instance metadata options](https://docs.aws.amazon.com/AWSEC2/latest/UserGuide/configuring-instance-metadata-options.html) in the *Amazon EC2 User Guide*. **Metadata token response hop limit** You can set the allowable number of network hops for the metadata token. If you do not specify a value, the default is 1. For more information, see [Configure the instance metadata options](https://docs.aws.amazon.com/AWSEC2/latest/UserGuide/configuring-instance-metadata-options.html) in the *Amazon EC2 User Guide*. **User data** You can customize and finish configuring your instances at launch time by specifying shell scripts or cloud-init directives as user data. The user data runs when the instance initially starts up, allowing you to automatically install applications, dependencies, or customizations at launch time. For more information, see [Run commands on your Linux instance at launch](https://docs.aws.amazon.com/AWSEC2/latest/UserGuide/user-data.html) in the *Amazon EC2 User Guide*. If you have large downloads or complex scripts, this adds to the time it takes for the instance to become ready for use. In which case, you may need to configure a lifecycle hook to delay an instance from reaching the `InService` state until it's fully provisioned. For more information about adding a lifecycle hook to your Auto Scaling group, see [Amazon EC2 Auto Scaling lifecycle hooks](lifecycle-hooks.md). # Request Spot Instances for fault-tolerant and flexible applications Request Spot Instances In your launch template, you can optionally request Spot Instances with no end date or duration. Amazon EC2 Spot Instances are spare capacity available at steep discounts compared to the EC2 On-Demand price. Spot Instances are a cost-effective choice if you can be flexible about when your applications run and if your applications can be interrupted. For more information about creating a launch template that requests Spot Instances, see [Create a launch template using advanced settings](advanced-settings-for-your-launch-template.md). **Important** Spot Instances are typically used to supplement On-Demand Instances. For this scenario, you can specify the same settings that are used to launch Spot Instances as part of the settings of your Auto Scaling group. When you specify the settings as part of the Auto Scaling group, you can request to launch Spot Instances only after launching a certain number of On-Demand Instances and then continue to launch some combination of On-Demand Instances and Spot Instances as the group scales. For more information, see [Auto Scaling groups with multiple instance types and purchase options](ec2-auto-scaling-mixed-instances-groups.md). This topic describes how to launch only Spot Instances in your Auto Scaling group by specifying settings in a launch template, rather than in the Auto Scaling group itself. The information in this topic also applies to Auto Scaling groups that request Spot Instances with a [launch configuration](create-launch-config.md). The difference is that a launch configuration requires a maximum price, but for launch templates, the maximum price is optional. When you create a launch template to launch only Spot Instances, keep the following considerations in mind: + **Spot price.** You pay only the current Spot price for the Spot Instances that you launch. This pricing changes slowly over time based on long-term trends in supply and demand. For more information, see [Spot Instances](https://docs.aws.amazon.com/AWSEC2/latest/UserGuide/using-spot-instances.html) and [Pricing and savings](https://docs.aws.amazon.com/AWSEC2/latest/UserGuide/using-spot-instances.html#spot-pricing) in the *Amazon EC2 User Guide*. + **Setting your maximum price.** You can optionally include a maximum price per hour for Spot Instances in your launch template. If your maximum price exceeds the current Spot price, the Amazon EC2 Spot service fulfills your request immediately if capacity is available. If the price for Spot Instances rises above your maximum price for a running instance in your Auto Scaling group, it terminates your instance. **Warning** Your application might not run if you do not receive any Spot Instances, such as when your maximum price is too low. To take advantage of the Spot Instances available for as long as possible, set your maximum price close to the On-Demand price. + **Balancing across Availability Zones.** If you specify multiple Availability Zones, Amazon EC2 Auto Scaling distributes the Spot requests across the specified zones. If your maximum price is too low in one Availability Zone for any requests to be fulfilled, Amazon EC2 Auto Scaling checks whether requests were fulfilled in the other zones. If so, Amazon EC2 Auto Scaling cancels the requests that failed and redistributes them across the Availability Zones that have requests fulfilled. If the price in an Availability Zone with no fulfilled requests drops enough that future requests succeed, Amazon EC2 Auto Scaling rebalances across all of the Availability Zones. + **Spot Instance termination.** Spot Instances can be terminated at any time. The Amazon EC2 Spot service can terminate Spot Instances in your Auto Scaling group as the availability of, or price for, Spot Instances changes. When scaling or performing health checks, Amazon EC2 Auto Scaling can also terminate Spot Instances in the same way that it can terminate On-Demand Instances. When an instance is terminated, any storage is deleted. + **Maintaining your desired capacity.** When a Spot Instance is terminated, Amazon EC2 Auto Scaling attempts to launch another Spot Instance to maintain the desired capacity for the group. If the current Spot price is less than your maximum price, it launches a Spot Instance. If the request for a Spot Instance is unsuccessful, it keeps trying. + **Changing your maximum price.** To change your maximum price, create a new launch template or update an existing launch template with the new maximum price, and then associate it with your Auto Scaling group. The existing Spot Instances continue to run as long as the maximum price specified in the launch template used for those instances is higher than the current Spot price. If you did not set a maximum price, the default maximum price is the On-Demand price. # Use Capacity Blocks for machine learning workloads Capacity Blocks for ML Capacity Blocks help you reserve highly sought-after GPU instances on a future date to support your short-duration, machine learning (ML) workloads. For an overview of Capacity Blocks and how they work, see [Capacity Blocks for ML](https://docs.aws.amazon.com/AWSEC2/latest/UserGuide/ec2-capacity-blocks.html) in the *Amazon EC2 User Guide*. To start using Capacity Blocks, you create a capacity reservation in a specific Availability Zone. Capacity Blocks are delivered as `targeted` capacity reservations in a single Availability Zone. When you create your launch template, specify the Capacity Block's reservation ID and instance type. Then, update your Auto Scaling group to use the launch template you created and the Capacity Block's Availability Zone. When your Capacity Block reservation begins, use scheduled scaling to launch the same number of instances as your Capacity Block reservation. **Important** Capacity Blocks are only available for certain Amazon EC2 instance types and AWS Regions. For more information, see [Prerequisites](https://docs.aws.amazon.com/AWSEC2/latest/UserGuide/capacity-blocks-using.html#capacity-blocks-prerequisites) in the *Amazon EC2 User Guide*. **Topics** + [ ## Operational guidelines ](#capacity-blocks-operational-guidelines) + [ ## Specify a Capacity Block in your launch template ](#specify-a-capacity-block-in-your-launch-template) + [ ## Limitations ](#capacity-blocks-limitations) + [ ## Related resources ](#capacity-blocks-related-resources) ## Operational guidelines The following are basic operational guidelines that you should follow when using a Capacity Block with an Auto Scaling group. + Scale in your Auto Scaling group to zero more than 30 minutes before the Capacity Block reservation end time. Amazon EC2 will terminate any instances that are still running 30 minutes before the end time of the Capacity Block. + We recommend that you use scheduled scaling to scale out (add instances) and scale in (remove instances) at the appropriate reservation times. For more information, see [Scheduled scaling for Amazon EC2 Auto Scaling](ec2-auto-scaling-scheduled-scaling.md). + Add lifecycle hooks as needed to perform a graceful shutdown of your application inside the instances when scaling in. Leave enough time for the lifecycle action to complete *before* Amazon EC2 starts forcibly terminating your instances 30 minutes before the Capacity Block reservation end time. For more information, see [Amazon EC2 Auto Scaling lifecycle hooks](lifecycle-hooks.md). + Make sure that the Auto Scaling group points to the correct version of the launch template for the entire duration of the reservation. We recommend pointing to a specific version of the launch template instead of the `$Default` or `$Latest` version. **Note** If you leave a Capacity Block instance running until the end of the reservation and Amazon EC2 reclaims it, the scaling activities for your Auto Scaling group state that it was "`taken out of service in response to an EC2 health check that indicated it had been terminated or stopped`", even though it was purposely reclaimed at the end of the Capacity Block. Similarly, Amazon EC2 Auto Scaling will attempt to replace the instance in the same manner as it does for any instance that fails a health check. For more information, see [Health checks for instances in an Auto Scaling group](ec2-auto-scaling-health-checks.md). ## Specify a Capacity Block in your launch template To create a launch template that targets a specific Capacity Block for your Auto Scaling group, use one of the following methods: ------ #### [ Console ] **To specify a Capacity Block in your launch template (console)** 1. Open the Amazon EC2 console at [https://console.aws.amazon.com/ec2/](https://console.aws.amazon.com/ec2/). 1. On the top navigation bar, select the AWS Region where you created your Capacity Block. 1. On the navigation pane, under **Instances**, choose **Launch Templates**. 1. Choose **Create launch template**, and create the launch template. Include the ID of the Amazon Machine Image (AMI), the instance type, and any other launch template settings as needed. 1. Expand the **Advanced details** section to view the advanced settings. 1. For **Purchasing option**, choose **Capacity Blocks**. 1. For **Capacity reservation**, choose **Target by ID**, and then for **Capacity reservation - Target by ID**, choose the capacity reservation ID of an existing Capacity Block. 1. When you have finished, choose **Create launch template**. For help creating an Auto Scaling group with a launch template, see [Create an Auto Scaling group using a launch template](create-asg-launch-template.md). ------ #### [ AWS CLI ] **To specify a Capacity Block in your launch template (AWS CLI)** Use the following [create-launch-template](https://awscli.amazonaws.com/v2/documentation/api/latest/reference/ec2/create-launch-template.html) command to create a launch template that specifies an existing Capacity Block reservation ID. Replace each *user input placeholder* with your own information. ``` aws ec2 create-launch-template --launch-template-name my-template-for-capacity-block \ --version-description AutoScalingVersion1 --region us-east-2 \ --launch-template-data file://config.json ``` **Tip** If this command throws an error, make sure that you have updated the AWS CLI locally to the latest version. Contents of `config.json`. ``` { "ImageId": "ami-04d5cc9b88example", "InstanceType": "p4d.24xlarge", "SecurityGroupIds": [ "sg-903004f88example" ], "KeyName": "MyKeyPair", "InstanceMarketOptions": { "MarketType": "capacity-block" }, "CapacityReservationSpecification": { "CapacityReservationTarget": { "CapacityReservationId": "cr-02168da1478b509e0" } } } ``` The following is example output. ``` { "LaunchTemplate": { "LaunchTemplateId": "lt-068f72b724example", "LaunchTemplateName": "my-template-for-capacity-block", "CreateTime": "2023-10-27T15:12:44.000Z", "CreatedBy": "arn:aws:iam::123456789012:user/Bob", "DefaultVersionNumber": 1, "LatestVersionNumber": 1 } } ``` You can use the following [describe-launch-template-versions](https://awscli.amazonaws.com/v2/documentation/api/latest/reference/ec2/describe-launch-template-versions.html) command to verify the Capacity Block reservation ID associated with the launch template. ``` aws ec2 describe-launch-template-versions --launch-template-names my-template-for-capacity-block \ --region us-east-2 ``` The following is example output for a launch template that specifies a Capacity Block reservation. ``` { "LaunchTemplateVersions": [ { "LaunchTemplateId": "lt-068f72b724example", "LaunchTemplateName": "my-template-for-capacity-block", "VersionNumber": 1, "CreateTime": "2023-10-27T15:12:44.000Z", "CreatedBy": "arn:aws:iam::123456789012:user/Bob", "DefaultVersion": true, "LaunchTemplateData": { "ImageId": "ami-04d5cc9b88example", "InstanceType": "p5.48xlarge", "SecurityGroupIds": [ "sg-903004f88example" ], "KeyName": "MyKeyPair", "InstanceMarketOptions": { "MarketType": "capacity-block" }, "CapacityReservationSpecification": { "CapacityReservationTarget": { "CapacityReservationId": "cr-02168da1478b509e0" } } } } ] } ``` ------ ## Limitations + Support for Capacity Blocks is only available if your Auto Scaling group has a compatible configuration. Mixed instances groups and warm pools are not supported. + You can only target one Capacity Block at a time. ## Related resources + For the prerequisites and recommendations for using P5 Instances, see [Get started with P5 instances](https://docs.aws.amazon.com/AWSEC2/latest/UserGuide/p5-instances-started.html) in the *Amazon EC2 User Guide*. + Amazon EKS supports using Capacity Blocks to support your short duration, machine learning (ML) workloads on Amazon EKS clusters. For more information, see [Capacity Blocks for ML](https://docs.aws.amazon.com/eks/latest/userguide/capacity-blocks.html) in the **Amazon EKS User Guide**. + You can use Capacity Blocks with supported instance types and Regions. However, On-Demand Capacity Reservations provide flexibility to reserve capacity for other instances types and Regions. For a tutorial that shows you how to use the On-Demand Capacity Reservation option, see [Reserve capacity in specific Availability Zones with Capacity Reservations](use-ec2-capacity-reservations.md).