OpenResty, PHP-FPM, MariaDB and Redis on 100% NVMe. Operated by us, in Germany. No cPanel, no Plesk, no license overhead in your plan.
No cPanel or Plesk fees baked into your plan. The money that would have gone to a control-panel renewal goes into hardware, monitoring and people.
OpenResty handles the edge. PHP-FPM runs your app. MariaDB stores it. Redis caches it. Each layer is documented, swappable and inspectable.
Patching, monitoring, backups and on-call live with us. You ship code. We keep the floor stable. The boundary between us and your work is clear.
We coordinate the move — files, database, email, DNS — scoped with you and validated on staging before cutover. If something breaks during a migration we run, that is ours to fix, not yours.
Standard NGINX reads config. OpenResty runs logic — rate limiting, inline WAF — directly in the request lifecycle. On Managed Premium, this is part of the platform, not an add-on.
Start on shared hosting, move to VPS, move to Managed Premium, move to a fully operated application platform. Same provider, same engineering logic, no forced rebuilds.
cPanel costs around €45 per server per month, and Plesk is not free either — every platform that wraps the server in a proprietary UI passes that cost on, in the price or in slower response times.
We made a different decision: our stack is NGINX or OpenResty, PHP-FPM, MariaDB and Redis — open source, tuned directly, no license renewal required to keep it running.
OpenResty edge, PHP-FPM per-site pool, MariaDB and Redis tuned for typical PHP workloads. NVMe storage by default. Suitable for sites and small apps where you do not need a full VPS.
Redis object cache, edge cache that bypasses cart and checkout, WP-CLI on every plan. Configured for store traffic, not generic PHP. Migrations handled by us if you want them handled.
Managed Cloud VPS with root access included. YNVAR configures, monitors and maintains the platform layer; you own application code and content. Hetzner for cost-per-core, UpCloud for IO-bound workloads. Both inside the EU.
You get the application configured, monitored, backed up and patched. We hold the on-call rotation. You do not maintain the server, you do not chase upgrades. Use the platform.
Answer a few questions — we route you to the product path that fits, with a starting price when one applies.
Pick the closest match — we tune the recommendation to the workload.
Hosting only = you ship the app. Managed ops = we run the platform layer. Managed app = we run the application too.
Travels with “Talk to engineering” — not stored in the URL.
A small interface, not a feature wall. Switch between hosting, VPS, backups, deployments, and domains — every screen reflects what is actually running, never a sales mock.
The Linux platform underneath every YNVAR product is the same — OpenResty, PHP-FPM, MariaDB / PostgreSQL, Redis, NVMe. What changes is the operating model: how much of the stack you run versus how much YNVAR operates. Pick the application; each card routes to the right model.
Sites & stores
Routed to Smart Web, Enterprise, or Managed Premium by workload size.
Operated applications
Operated end-to-end by YNVAR, with on-call ownership of the runtime.
Frameworks · runtimes · automation
Run on a Managed Cloud VPS (root) or Managed Premium (operated).
PHP framework
JS runtime
Workflow automation
AI workflow
Self-service deploy
The platform under your application is monitored, patched and backed up by us — and the boundary between our work and yours is documented before you sign.
Daily snapshots with 24h / 7d / 30d retention buckets. Stored independently of the production host so a hardware loss does not take the backup with it.
Every layer of the stack — edge, runtime, database, cache, storage — reports into the same pipeline. If a service degrades, our on-call sees it before you do.
Backups are only as good as the last successful restore. We exercise the path regularly on test environments so the operation is not theoretical the day you need it.
Where YNVAR ends and your app begins is written down. We run the platform — OS, web server, runtime, database, cache, network. You ship code and content.
// Illustrative diagrams. No claimed uptime, no SLA promises invented for marketing.
Shared hosting, VPS, or managed application hosting — same engineering logic, one provider, no forced rebuild.
Operated from Germany · Hetzner · UpCloud · Backups & on-call on managed plans.
