Terraform vs Pulumi vs CloudFormation: IaC Comparison

In this tutorial, you'll learn about Terraform vs Pulumi vs CloudFormation: IaC Comparison. We cover key concepts, practical examples, and best practices to help you understand and apply this topic effectively.

Terraform, Pulumi, and CloudFormation are the leading Infrastructure as Code (IaC) tools — Terraform uses HCL a declarative DSL, Pulumi lets you write infrastructure in real programming languages, and CloudFormation provides AWS-native template-based provisioning.

At a Glance

Key Differences

• Language: Terraform's HCL is declarative — you describe desired state in resource "aws_s3_bucket" "my_bucket" { ... }. Pulumi uses TypeScript, Python, Go, C#, or Java — you write infrastructure using familiar language constructs (loops, conditionals, classes). CloudFormation uses JSON or YAML with intrinsic functions (Fn::Join, Ref, !Sub).
• State Management: Terraform stores state in a backend (S3, Terraform Cloud) — state files contain resource metadata and can be locked to prevent concurrent modifications. Pulumi uses managed state via Pulumi Cloud (free for individuals) or self-managed backends. CloudFormation manages state on AWS — no manual state files, but limited to AWS.
• Testing: Pulumi's use of real programming languages enables Unit Testing with your existing test framework. Terraform requires specialized tools like Terratest or OPA policies. CloudFormation Guard provides policy-as-code validation.
• Multi-Cloud: Terraform and Pulumi both support AWS, Azure, GCP, Kubernetes, and hundreds of other providers. CloudFormation only works with AWS. If you run a Multi-Cloud or hybrid infrastructure, Terraform or Pulumi are necessary.

When to Choose Terraform

Terraform is the industry standard for IaC — the largest provider ecosystem, the most community modules, and the widest adoption. HCL may require learning a new language, but its declarative syntax maps perfectly to infrastructure. Terraform's plan/apply workflow gives clear visibility into changes before they're applied. The Terraform Registry (3,000+ modules) lets you reuse battle-tested configurations for VPCs, Kubernetes clusters, and databases. Terraform works everywhere — AWS, Azure, GCP, Kubernetes, Helm, GitHub, and on-premises.

When to Choose Pulumi

Pulumi is best for teams who want to use their existing programming language expertise for infrastructure. If your team knows TypeScript, you can write infrastructure with the same IDE, linters, Type Checking, and testing patterns as your application code. Pulumi excels in complex scenarios — loops create multiple resources, conditionals control resource creation, and shared code (npm packages) distributes infrastructure patterns across teams. Pulumi's pulumi up CLI provides a similar workflow to Terraform but with richer automation possibilities through native language constructs.

When to Choose CloudFormation

CloudFormation is the best choice for AWS-only workloads where you want native integration, automatic drift detection, and no state files to manage. CloudFormation StackSets let you deploy across multiple accounts and regions with built-in permission management. CloudFormation's drift detection runs automatically and reports resource changes. For organizations that are all-in on AWS and need minimal tooling overhead, CloudFormation provides a seamless experience within the AWS console.

Architecture Comparison

flowchart TD
    subgraph Terraform
        T1[HCL Config] --> T2[terraform plan]
        T2 --> T3[State File]
        T3 --> T4[AWS/Azure/GCP]
    end
    subgraph Pulumi
        P1[TypeScript/Python] --> P2[pulumi up]
        P2 --> P3[Pulumi Cloud State]
        P3 --> P4[AWS/Azure/GCP]
    end
    subgraph CloudFormation
        C1[YAML/JSON] --> C2[AWS CloudFormation]
        C2 --> C3[AWS Drift Detection]
        C3 --> C4[AWS Resources]
    end

Side by Side Code Example: Create an S3 Bucket

Terraform (HCL)

resource "aws_s3_bucket" "my_bucket" {
  bucket = "my-app-data-2026"
  tags = {
    Name        = "My App Bucket"
    Environment = "production"
  }
}

resource "aws_s3_bucket_versioning" "my_bucket" {
  bucket = aws_s3_bucket.my_bucket.id
  versioning_configuration {
    status = "Enabled"
  }
}

Pulumi (TypeScript)

import * as aws from "@pulumi/aws";

const bucket = new aws.s3.Bucket("my-bucket", {
  bucket: "my-app-data-2026",
tags: {
    Name: "My App Bucket",
    Environment: "production",
  },
});

new aws.s3.BucketVersioningV2("my-bucket-versioning", {
  bucket: bucket.id,
  versioningConfiguration: {
    status: "Enabled",
  },
});

export const bucketName = bucket.id;

CloudFormation (YAML)

Resources:
  MyBucket:
    Type: AWS::S3::Bucket
    Properties:
      BucketName: my-app-data-2026
      Tags:
        - Key: Name
          Value: My App Bucket
        - Key: Environment
          Value: production
  MyBucketVersioning:
    Type: AWS::S3::BucketVersioning
    Properties:
      Bucket: !Ref MyBucket
      VersioningConfiguration:
        Status: Enabled

All three create an S3 bucket with versioning. Terraform uses HCL's resource blocks. Pulumi uses TypeScript with new aws.s3.Bucket(). CloudFormation uses YAML with !Ref intrinsic functions. Pulumi's code is plain TypeScript — you get IDE autocompletion, Type Checking, and the ability to use loops and conditionals.

FAQ

{{< faq "Can Pulumi use Terraform providers?">}} Yes. Pulumi supports Terraform providers via the Pulumi Terraform Bridge. This means Pulumi can manage any resource available in the Terraform ecosystem, giving it access to 2000+ providers while using real programming languages. {{< /faq >}}

{{< faq "Is Terraform still free?">}} Terraform itself is open source and free. Terraform Cloud has a free tier with limitations (5 users, 500 resources). Additional features (team management, policy enforcement, private module registry) require paid plans. {{< /faq >}}

{{< faq "Can I migrate from CloudFormation to Terraform?">}} Yes. Terraform's <a href="/devops/terraform/">Terraform</a> import can import existing CloudFormation-managed resources. Tools like tfmigrate and terraformer automate migration. The process involves importing state, writing HCL config, and then managing the stack purely in Terraform. {{< /faq >}}