Edge infrastructure management
AWS provides fully managed services that extend AWS infrastructure, services, APIs,
and tools closer to your end users and data centers. The services that are available in
Outposts and Local Zones are the same as those available in AWS Regions, so you can manage those
services by using the same AWS console, AWS CLI, or AWS APIs. For supported services, see
the AWS Outposts feature comparison
Deploying services at the edge
You can configure the available services in Local Zones and Outposts in the same way you configure them in AWS Regions: by using the AWS console, AWS CLI, or AWS APIs. The primary difference between Regional and edge deployments is the subnets where resources will be provisioned. The Networking at the edge section described how subnets are deployed in Outposts and Local Zones. After you identify the edge subnets, you use the edge subnet ID as a parameter to deploy the service in Outposts or Local Zones. The following sections provide examples of deploying edge services.
Amazon EC2 at the edge
The following run-instances
example launches a single instance of
type m5.2xlarge
into the edge subnet for the current Region. The key
pair is optional if you do not plan to connect to your instance by using SSH on
Linux or remote desktop protocol (RDP) on Windows.
aws ec2 run-instances \ --image-id ami-id \ --instance-type m5.2xlarge \ --subnet-id <subnet-edge-id> \ --key-name MyKeyPair
Application Load Balancers at the edge
The following create-load-balancer
example creates an internal Application Load Balancer
and enables the Local Zones or Outposts for the specified subnets.
aws elbv2 create-load-balancer \ --name my-internal-load-balancer \ --scheme internal \ --subnets <subnet-edge-id>
To deploy an internet-facing Application Load Balancer to a subnet on an Outpost, you set the
internet-facing
flag in the --scheme
option and
provide a CoIP pool
ID, as shown in this example:
aws elbv2 create-load-balancer \ --name my-internal-load-balancer \ --scheme internet-facing \ --customer-owned-ipv4-pool <coip-pool-id> --subnets <subnet-edge-id>
For information about deploying other services at the edge, follow these links:
Service |
AWS Outposts |
AWS Local Zones |
---|---|---|
Amazon EKS |
||
Amazon ECS |
Amazon ECS applications in shared subnets, Local Zones, and Wavelength Zones |
|
Amazon RDS |
Select the Local Zone subnet |
|
Amazon S3 |
Not available |
|
Amazon ElastiCache |
||
Amazon EMR |
||
Amazon FSx |
Not available |
Select the Local Zone subnet |
AWS Elastic Disaster Recovery |
Not available |
|
AWS Application Migration Service |
Not available |
Select the Local Zone subnet as the staging subnet |
Outposts-specific CLI and SDK
AWS Outposts has two groups of commands and APIs for creating a service order or manipulating the routing tables between the local gateway and your local network.
Outposts ordering process
You can use the AWS CLI
Local gateway management
The management and operation of the local gateway (LGW) in Outposts requires knowledge of the AWS CLI and SDK commands available for this task. You can use the AWS CLI and AWS SDKs to create and modify LGW routes, among other tasks. For more information about managing the LGW, see these resources:
-
EC2.Client in the AWS SDK for Python (Boto)
-
Ec2Client in the AWS SDK for Java
CloudWatch metrics and logs
For AWS services that are available in both Outposts and Local Zones, metrics and logs are managed in the same way as in Regions. Amazon CloudWatch provides metrics that are dedicated to monitoring Outposts in the following dimensions:
Dimension |
Description |
---|---|
|
The account or service using the capacity |
|
The instance family |
|
The instance type |
|
The ID of the Outpost |
|
The EBS volume type |
|
The ID of the local gateway or service link virtual interface (VIF) |
|
The ID of the VIF group for the local gateway VIF |
For more information, see CloudWatch metrics for Outposts racks in the Outposts documentation.