From AWS EKS to AWS EKS with Qovery - 6 Reasons Why This Startup Migrated

Sometimes, even when you’re running on a solid infrastructure like AWS EKS, a growing company needs to rethink how they manage it. One startup, fresh off a successful Series A raise, found themselves in just that situation. They had been running two EKS clusters — Production and Staging — managed entirely by their CTO, who is an experienced AWS power user. But with the company scaling and a larger software engineering team on the horizon, they faced a crucial decision: hire their first DevOps Engineer or delay that hire by using Qovery. Here’s why they chose the latter.

Romaric Philogène

Romaric Philogène

November 11, 2024 · 4 min read
From AWS EKS to AWS EKS with Qovery - 6 Reasons Why This Startup Migrated - Qovery

#6 Reasons Why This Startup Migrated From EKS to EKS with Qovery

#1. No More EKS Upgrade

Despite being an AWS expert, the CTO wanted to free up time for more strategic work and ensure the infrastructure remained best-in-class. Qovery, as an ISV AWS Accelerate Partner, fit the bill perfectly. Qovery’s managed EKS service takes care of everything: cluster upgrades, patch management, and overall maintenance.

Qovery Frees Up 100% of the Setup and Maintenance Time Required to Operate AWS EKS (Kubernetes)
Qovery Frees Up 100% of the Setup and Maintenance Time Required to Operate AWS EKS (Kubernetes)

By choosing Qovery, the startup got a state-of-the-art infrastructure running on AWS, without the hassle of managing it themselves. It’s like having a fully automated, always up-to-date EKS setup handled by a dedicated team.

Qovery can also be installed on their existing EKS cluster, but in that case, we can't provide day-2 management (yet). So, their preferred choice was to migrate to our managed EKS cluster running on their own cloud account.

#2. Give Autonomy To Software Engineers

As the software engineering team grew, the CTO needed to strike a delicate balance: give developers the autonomy to deploy and manage their own services while maintaining control and security. Qovery’s Role-Based Access Control (RBAC) and Audit Logs made this possible. Software engineers could operate independently without risking security compliance or governance standards, and the CTO could still have oversight whenever needed. Qovery’s tools ensured that security wasn’t compromised even as developer autonomy increased.

#3. Best-in-Class IaC

For this CTO, Infrastructure-as-Code (IaC) was non-negotiable. Having previously managed their AWS infrastructure using Terraform, they needed any new solution to seamlessly integrate with their IaC workflows. They were sold when they discovered that Qovery had an official Terraform Provider.

resource "qovery_application" "my_app" {
  environment_id = qovery_environment.production.id
  name           = "My App"
  cpu            = 500 # Allocate 0.5 CPU
  memory         = 512 # Allocate 512 MB of RAM
  git_repository = {
    url       = "https://github.com/evoxmusic/my-app.git" # Qovery will clone and build the app directly
    branch    = "main"
    root_path = "/"
  }
  build_mode            = "DOCKER"
  dockerfile_path       = "Dockerfile" # It will uses this Dockerfile
  min_running_instances = 1
  max_running_instances = 3 # Scale up to 3 instances
  ports = [
    {
      # Expose this app publicly on HTTPS and provide TLS certificate automatically
      internal_port       = 1337
      external_port       = 443
      protocol            = "HTTP"
      publicly_accessible = true
      is_default          = true
    }
  ]
  environment_variables = [
    {
      key   = "DATABASE_HOST"
      value = qovery_database.my_psql_database.internal_host
    },
    {
      key   = "DATABASE_PORT"
      value = qovery_database.my_psql_database.port
    },
    {
      key   = "DATABASE_USERNAME"
      value = qovery_database.my_psql_database.login
    },
    {
      key   = "DATABASE_NAME"
      value = "postgres"
    },
  ]
  secrets = [
    {
      key   = "ADMIN_JWT_SECRET"
      value = var.strapi_admin_jwt_secret
    },
    {
      key   = "API_TOKEN_SALT"
      value = var.strapi_api_token_salt
    },
    {
      key   = "APP_KEYS"
      value = var.strapi_app_keys
    },
    {
      key   = "DATABASE_PASSWORD"
      value = qovery_database.my_psql_database.password
    }
  ]
  healthchecks = {
    ...
  }
}

Everything that could be configured through Qovery’s web interface could also be automated and managed via Terraform. This meant the company could continue leveraging their existing Terraform expertise while taking advantage of Qovery’s streamlined infrastructure management.

#4. Effortless Future Scaling

Scalability was another key factor. The company wanted to keep open the option of spinning up additional EKS clusters as they expanded, without worrying about the maintenance and complexity that comes with it.

Create Multiple Clusters and Unified The Experience
Create Multiple Clusters and Unified The Experience

Qovery’s platform made it easy to launch and manage new clusters in the future. With Qovery handling the heavy lifting, adding new clusters won’t require additional DevOps resources, saving both time and money.

#5. 10x Cheaper

The final, and perhaps most compelling, reason boiled down to cost. Hiring a full-time DevOps Engineer in NYC, with a complete package, would have set them back around $220k per year. By comparison, using Qovery was roughly 10 times less expensive.

The choice was obvious: Qovery offered a cost-effective alternative that allowed the company to delay their DevOps hire while still scaling their infrastructure seamlessly. For a startup looking to be efficient with their funding, it was a no-brainer.

#6. Hassle-Free Migration

One of the reasons this migration was so fast and smooth was because the company was already using Helm Charts to manage their Kubernetes workloads. Qovery not only supports Helm Charts natively but also takes their management to the next level.

With Qovery, the company could easily leverage environment variables and secrets directly in their Helm deployments, simplifying configuration management. Additionally, Qovery offers native GitOps support for Helm values files, allowing them to version and manage their infrastructure configurations alongside their application code. This streamlined their workflows and ensured consistency across environments.

Thanks to Qovery’s seamless integration with Helm, the entire migration process was completed quickly, allowing the team to focus on delivering value rather than troubleshooting infrastructure issues.

#So What?

Migrating from AWS EKS to AWS EKS with Qovery might seem unconventional at first, but for this growing startup, it made perfect sense. By using Qovery, they maintained their AWS-native setup, empowered their software engineers, guaranteed infrastructure security, and achieved significant cost savings. The transition has not only prepared them for future growth but also freed up their CTO to focus on innovation rather than maintenance.

If you’re in a similar position, facing the challenge of scaling infrastructure efficiently while keeping costs down, it might be worth exploring a solution like Qovery. Sometimes, the best upgrade isn’t about switching platforms but optimizing how you use the one you already have.

Your Favorite DevOps Automation Platform

Qovery is a DevOps Automation Platform Helping 200+ Organizations To Ship Faster and Eliminate DevOps Hiring Needs

Try it out now!
Your Favorite DevOps Automation Platform
Qovery white logo

Your Favorite DevOps Automation Platform

Qovery is a DevOps Automation Platform Helping 200+ Organizations To Ship Faster and Eliminate DevOps Hiring Needs

Try it out now!
AWSKubernetes