FRENIT
24/7 emergency

Cloud-native

Kubernetes & cloud-native

Your databases, orchestrated. Without sacrificing reliability.

Stateful workloads on Kubernetes are rightly intimidating: a poorly operated database inside a cluster means an incident that spreads fast. We combine our two areas of expertise — the heart of data and systems engineering — to design and operate cloud-native platforms where your databases run in production with confidence. And because we run what we sell, we only talk about what we actually practise.

Cluster design & operations

Build a platform meant to last in production. Cluster architecture, sizing, resilience and upgrade strategies, hardening — from a starter cluster to a multi-node platform operated day to day.

  • Kubernetes cluster architecture and sizing
  • High availability, upgrade and recovery strategies
  • Hardening, access management and best practices

Stateful databases on Kubernetes

Let data and orchestration coexist, with confidence. Deployment and operation of databases through proven operators — CloudNativePG for PostgreSQL, Percona Operators, MongoDB Operator — with the backup, failover and restore rigour that has been our craft from the start.

  • Database operators (CloudNativePG, Percona, MongoDB)
  • Automated backup, replication and failover
  • Migration of existing databases to Kubernetes

GitOps & industrialisation

Deploy in a reproducible, traceable, reversible way. GitOps pipelines (Argo CD, Flux), Infrastructure as Code and deployment pipelines, so every change is versioned, reviewed and revertible.

  • GitOps workflows (Argo CD, Flux)
  • Continuous, reproducible deployment pipelines
  • Versioned configuration management

Observability & monitoring

See what is happening before it breaks. Metrics, logs and alerting with Prometheus and Grafana — which we can plug into our own DBK platform for a unified view of your databases and systems.

  • Prometheus / Grafana monitoring
  • Alerting and operational dashboards
  • Integration with the DBK platform

A production cluster needs watching and support. Our 24/7 on-call naturally extends to your Kubernetes platforms.