Back to blog

Generating a Simulation from Terraform

Agent-based simulations of IaaC
May 26th, 2021
Generating a Simulation from Terraform

The Infrastructure-as-Code (IaC) movement has revolutionized the way cloud infrastructure is managed, enabling self-documenting, version-controlled provisioning and maintenance of systems and hardware. While sysadmins are very much still required, tooling and best practices for programmatically setting up large computing infrastructures have dramatically improved as DevOps has matured. In particular, planning tools like Terraform make it easy to set up and maintain massive resource clusters with relative ease.

Because infrastructure is defined in declarative information files, those same configurations can be used as inputs within HASH to automatically generate models that can be used for simulation.

Terraform-based HASH simulations can help improve infrastructure provisioning and design processes, helping you select the right instance types to use, anticipate or avoid downtime, and forecast or smooth costs.

By way of lightweight demonstration, above you'll see a simulation that takes in a Terraform configuration alongside a set of user data, and creates a Kubernetes cluster. This leverages HASH's support for datasets to define a unique simulation based off of the user's configuration files:

  • terraform_resource.json - a style file that defines the cluster and gets information on AWS CPU and memory specs from another dataset of AWS instances.
  • distribution.csv - a dataset that represents a proportion of daily requests received each hour of the day.

Pods and instance agents contain behaviors that receive messages from a request generator. The requests are stochastically generated based on the data in distribution.csv. Read more in the project's README ->

Create a free account

Sign up to try HASH out for yourself, and see what all the fuss is about

By signing up you agree to our terms and conditions and privacy policy