Why run Quarkus here?
Our management platform Barista is built on Quarkus. We work with the framework updates, extensions and deployment questions that arise in production ourselves.
You speak directly with the engineers managing your environment. With Advanced, we also respond outside office hours to production P1 incidents: unavailability or data loss. See our response times and definitions.
Our own infrastructure in Dutch datacenters. We are working toward ISO 27001 certification.
Three ways to run your app
| Service | Who manages what | Monthly price |
|---|---|---|
| Pods — self-service | Your own namespace. You manage deployments and container images; we manage the cluster. | From €100 |
| Managed Pods | We manage the platform layer and managed database. You own the code, container images and releases. | From €350 including Cortado capacity and Advanced |
| Managed Servers (CaffeineStacks) | A dedicated VM environment with management of the OS, JVM, database engine and deployment mechanism. | From €300 with Basis; from €550 with Advanced |
Prices exclude VAT. Managed Pods from €350 covers one workload with a Cortado pod (4 GB, 2 vCPU, 75 GB), including its own database within that capacity. Need more workloads or capacity? See the bundles and worked examples.
Managed server example: an Espresso app server, Cortado database and Advanced total €650 per month (€300 + €100 + €250). Basis response times apply during office hours.
Taking over an existing application involves a one-time setup fee from €950. We agree the scope and price in advance. For larger environments, a dedicated managed cluster starts at €2,500 per month, with a €2,500 setup fee.
What we check during setup
- Memory and capacity: we size the environment around measured usage, including heap, native memory and the database.
- Health checks: we align startup, readiness and liveness checks with your application’s startup behaviour and dependencies.
- JVM / native: we support JVM and native deployments. The choice depends on your application, libraries and build process.
From our own work: loading a JDK 25 AOT cache on a different CPU.
Database
On Pods, you provision a database using the available operators. With Managed Pods, we take care of database management; an application with its own database counts as one workload. On CaffeineStacks, the database can run on the app VM or a separate database VM.
During a handover, we check the database version, connection, backups and recovery path. Database migrations or application code changes are agreed as separate work in advance.
Monitoring
Quarkus can expose metrics through Micrometer’s Prometheus extension. The monitoring system needs access to the metrics endpoint, with appropriate access settings.
During setup, we agree which metrics to collect, which dashboards your team needs and who receives each alert. These can include memory usage, GC pauses, response times and error rates. On self-service Pods, you configure the application side; for managed services, we agree the integration during onboarding.
Deployment
Your team decides what to release. During onboarding, we agree the build, registry, deployment and rollback process, using your existing CI/CD where possible. New pipelines and migration work have an agreed scope and price.
Operations and application work
For managed services, we investigate incidents even when the cause is in your application and provide a substantiated analysis. Bugfixes, application dependency updates and feature development fall outside platform management; we agree a separate fixed-price job for that work.
Your developer leaving? If you inherited a Quarkus application and soon no one can maintain it, we take over your Java application and keep it running.
