Terraform extends AWS support to cloud giant’s App Runner

Terraform extends AWS support to cloud giant’s App Runner

HashiCorp’s Terraform infrastructure-as-code provisioning tool has extended its ability to manage AWS resources. It now supports Amazon’s cloud arm’s new App Runner service and the new predictive scaling policy for EC2 instances.

Terraform is HashiCorp’s tool for managing IT resources, including public cloud and private cloud infrastructure, using declarative configurations. The firm has moved quickly to add support for the newly introduced AWS App Runner service into Terraform.

App Runner is a purpose-built container application service that allows users to build and run containerised web applications in a few clicks. It is powered by the AWS Fargate serverless compute platform. The latest version of the Terraform AWS provider provides support, while users must also have Terraform v0.12 or greater installed to take advantage of it.

To set up AWS App Runner in the Terraform AWS provider, users need to define some new resources. Find out how on the HashiCorp blog, which also has instructions on how to create an App Runner service built from a GitHub repository source.

Predictive scaling

HashiCorp also revealed the latest version of the Terraform AWS provider now supports predictive scaling policy for Amazon EC2 Auto Scaling. Predictive Scaling uses machine learning to analyse the history of the Auto Scaling group on AWS in order to predict future capacity requirements and provision additional EC2 instances in advance.

In order to use Predictive Scaling Policy in the Terraform AWS provider, users need to define one new resource – aws_predictive_scaling_configuration – as well as make some changes to the existing aws_autoscaling_policy resource. For more details on how to use the Terraform AWS provider to configure Predictive Scaling within an EC2 Auto Scaling group, see this post.

HashiCorp released Terraform 0.15 in April, stating it marked the beginning of the pre-release period that will lead up to the 1.0 release of the infrastructure-as-code provisioning tool. However, this should not be taken as a sign that the project is not sufficiently mature for production, HashiCorp insists.