Operational excellence on AWS: one EKS cluster, 67% lower costs

Quoted in this story:

JH

Jakub Handzus

AgeVolt

AgeVolt case study: Operational excellence on AWS: one EKS cluster, 67% lower costs

Intro

AgeVolt Slovakia s.r.o. is an innovative e-mobility startup headquartered in Bratislava, specializing in electric vehicle (EV) charging solutions. The company manages a rapidly expanding network of IoT-enabled charging stations, delivering smart infrastructure to support sustainable transportation across Europe. They partnered with us, DevOpsGroup s.r.o., to streamline their AWS infrastructure.

Project type: AWS Infrastructure Consolidation and DevOps Automation

Services: AWS Infrastructure Consolidation, DevOps Automation & GitOps (ArgoCD, CI/CD with GitHub Actions), Observability & Monitoring Setup (Grafana, Prometheus, Loki)

The challenge

Rapid growth led to fragmented infrastructure:

  • Multiple AWS accounts, increasing management complexity.
  • 4+ EKS clusters driving operational overhead.
  • Scattered DNS records hindering scalability.
  • High costs and risk of disruptions in service delivery.
  • Outdated Terraform code complicating maintenance.

These challenges aligned directly with the AWS DevOps Competency, emphasizing the need for consolidated cloud operations to handle high-availability IoT workloads effectively.

Our solution

This was a comprehensive consolidation project: we unified everything into a single AWS account and deployed a streamlined Amazon EKS cluster - replacing the previous 4+ clusters - with dedicated node pools for production, staging, shared and ephemeral services. Karpenter was integrated for dynamic autoscaling, automatically provisioning nodes to match workload demands and optimize compute efficiency. This approach minimized redundancy and ensured zero-downtime migration.

Key AWS services drove the transformation: Amazon RDS for MySQL provided managed, high-availability databases in production with automated backups, paired with Amazon ElastiCache for Redis to enable low-latency caching for real-time applications. Amazon Route53 centralized all DNS records, automated via External-DNS service to ensure seamless updates.

For IoT and messaging, EMQX ran on EKS for secure MQTTS communication with charging stations, while RabbitMQ handled internal queues. Observability was bolstered by integrating Grafana, Prometheus, Loki, InfluxDB, and Telegraf directly on AWS infrastructure for comprehensive monitoring.

DevOps practices were elevated with ArgoCD enabling GitOps for declarative deployments from Git repositories. The GitHub Actions Runner Controller (ARC), deployed as a Kubernetes operator on EKS, automated CI/CD by autoscaling ephemeral, container-based runners based on workflow needs - integrating natively with AWS to accelerate pipelines without extra infrastructure. Additional tools like external-secrets for integration with AWS Secrets Manager and ingress-nginx for Network Load Balancer provisioning rounded out the automated, resilient setup.

Results and benefits

By consolidating multiple accounts and clusters into a unified setup, we reduced monthly costs while boosting efficiency and scalability for their EV charging network. The AWS-powered consolidation delivered transformative outcomes for AgeVolt. Monthly costs plummeted by 67%, from approximately $4,500 to $1,500, thanks to reduced redundancies, optimized resource allocation via Karpenter, and streamlined operations in a single account. Deployment speeds increased with automated CI/CD, while centralized observability minimized downtime and enhanced system reliability.

Results at a glance

MetricBeforeAfterImprovement
AWS accountsMultiple1Simplified management
EKS clusters4+1Reduced overhead
Monthly cost~$4,500$1,50067% savings
DNS managementFragmentedCentralizedStreamlined updates
CI/CD efficiencyManual/FragmentedAutomated with ARCDynamic scaling & faster workflows

This transformation empowered AgeVolt to focus on business growth, not infrastructure woes.

Quote

We value our collaboration with the DevOpsGroup team very positively. Their approach is professional, and communication is efficient and prompt. We appreciate their reliability and consistency in task execution, as well as the team's readiness to respond to project needs. Overall, we perceive the cooperation as high-quality and beneficial, with a clear focus on achieving goals and ensuring customer satisfaction.
Jakub HandzusAgeVolt

Start a project like this one.

Tell us where your infrastructure is today and where it needs to be. We will map the path with you.

  • Zero-downtime migrations
  • GitOps delivery
  • Policy-as-code security
  • 24/7 platform SRE
  • Multi-cloud landing zones
  • Observability built in
  • Cost-aware scaling
  • Audit-ready compliance

Tell us about your project

Fill out the form and our team will get back to you shortly.

I agree to the Privacy Policy and consent to Axyra processing my data to answer this inquiry. *

Bring us the hard part.

Contact us