Full service release on Jan 1 • See details
ManoVersa Hosting

How our stack works

Platform overview

We operate on bare‑metal hosts with multi‑CPU configurations, ample RAM, and a mix of SSD/HDD storage. Services are containerized and monitored end‑to‑end.

Compute & isolation

Workloads run in containers/VMs to isolate resources cleanly while keeping performance predictable.

Game servers

Pterodactyl provides a modern panel for managing instances with sensible defaults and backup options.

Observability

Prometheus collects metrics and Grafana visualizes dashboards; we receive alerts for key thresholds.

We adopted a server. His name is Dell R630. – ManoVersa Blog
ManoVersa Hosting

Our dedicated Dell R630 setup

Monitoring: Prometheus + Grafana

Welcome to a behind-the-scenes look at our server infrastructure! In this post, we’ll walk you through our Dell R630 configuration, why we chose bare-metal, how we keep an eye on things, and what all this means for you.

What is a Dell R630?

The Dell PowerEdge R630 is a 1U rackmount server, known for its reliability and performance. Ours is equipped with dual Intel Xeon CPUs, 128GB of RAM, and a mix of SSDs and HDDs for both speed and storage capacity.

ELI5: What does "bare-metal" mean?

ELI5: Imagine renting a whole house (bare-metal) instead of just a room in an apartment building (the cloud). We have the keys, we control everything, and no one else is sharing the space.

Why do we use bare‑metal servers?

  • Performance: No noisy neighbors! All resources are dedicated to our customers.
  • Predictability: We know exactly what’s running and when.
  • Control: We decide when to update, reboot, or make changes—no surprises.
  • Security: Fewer layers means fewer things can go wrong.

How is it set up?

  • Storage: SSDs for fast access, HDDs for backups and archives.
  • Network: Gigabit NICs for redundancy and speed.
  • Power: Dual power supplies, connected to separate circuits for uptime.
  • Environment: The server lives in a cool, dry, and dust-free office (sorry, no petting allowed!).

ELI5: How do we know if something goes wrong?

ELI5: Think of Prometheus as a robot that checks the server’s “vitals” every few seconds—like temperature, speed, and if it’s feeling okay. If something looks weird, Grafana puts it on a big dashboard, and if it’s really bad, we get a message right away.

Our monitoring approach

  • Prometheus: Collects metrics (CPU, memory, disk, network, etc.) from the server and services.
  • Grafana: Visualizes all the data in real-time dashboards, so we can spot issues before they become problems.
  • Alerting: If something crosses a threshold (like high temperature or low disk space), we get notified instantly-day or night.

What does this mean for you?

  • Reliability: We catch issues early, so your services stay online.
  • Performance: Dedicated hardware means your apps run fast and smooth.
  • Transparency: We’re happy to share how things work under the hood!

Questions? Curious about our setup? Let us know! We love talking shop.