A system isn't stable because of what's inside it—it's stable because of what happens when things go wrong
The five points below describe the specific risks that Triox's infrastructure addresses. Every component runs in a configuration dedicated to each client—there are no prepackaged plans and no shared environment.
01Data loss and restore
For us, restore is not a plan—it's separate infrastructure
Most serious data loss isn't serious because it happened, but because the backup was in the same rack, on the same storage or in the same building as the production system.
Our backups sit on multiple, physically independent targets at multiple sites—and critical workloads also get an off-site VM replica, which you don't restore from but simply start.
The question isn't whether there is a backup. It's where it is, and how long it takes to be back up.
Technical details
Backup: enterprise-grade data backup solutions
Backup targets: multiple independent physical backup targets at multiple sites
Replication: off-site VM replica target for critical workloads
Data center: two independent sites in Budapest
02Client isolation
Your environment is not shared with anyone
In a shared environment, another client's compromise, misconfiguration or sudden load spike is your risk too—even if you did everything right.
With us, every client gets its own network segment, with VLAN-based or physical segmentation, behind a dedicated gateway. On request, the environment can be made completely unreachable from the internet—for clients handling sensitive data, this is already established practice.
In an audit, “logically separated” is often not enough. That's why we also offer physical segmentation.
Technical details
Client isolation: VLAN-based or physical segmentation, a dedicated segment per client
Gateway: a dedicated firewall, NAT and VPN endpoint per client—with its own rule set
Public IP: a dedicated public IPv4 address per client, from our own address range
Two-tier segmentation: separate internal (database, application) and public (web, mail) network layers
A dedicated IP from our own address range also means that the IP reputation of your email and services does not depend on what another client does.
03Management access
The attack surface starts where the management interface is
The security of an infrastructure is often decided not at the application, but by where the management interfaces of servers and network devices can be reached from.
Triox's management network is dedicated and not reachable from the internet—there is no exposed management interface that someone would have to protect with a password.
What can't be reached from the internet can't be hacked from the internet.
Technical details
Out-of-band management: a dedicated management network, not reachable from the internet
Client access via VPN, on a dedicated endpoint per client
04Performance and availability
The bottleneck is rarely the server
When a virtualized environment is slow, it is usually not a CPU issue but a network or storage one. Our backbone is 2×10 Gbps in a redundant design—losing an uplink or a switch does not mean a service outage.
Resource allocation follows your workload, not a prepackaged plan: we also handle enterprise-scale workloads with heavy compute and storage requirements.
Without redundancy, “high speed” lasts only as long as everything works.
Scalable resource allocation: CPU, memory and storage allocation tailored to your needs
No rigid package constraints: custom configuration, not prepackaged templates
05Virtualization platform
You don't have to switch platforms to move to us
Switching platforms is a project in itself: migration risk, new licenses, retraining, rewritten operational processes.
We run all three widely used platforms—Microsoft Hyper-V, VMware and Proxmox—so your existing environment can stay on whatever it runs on now. But if you are considering a switch precisely because of licensing costs, we can do that too, through a managed migration.
The platform should be a choice, not a constraint.
Technical details
Virtualization: Microsoft Hyper-V (Windows Server platform), VMware, Proxmox
Migration support from an existing environment
Service models: IaaS (VMs, storage, networking), DCaaS (a complete client data center), SaaS (application hosting)
Let's look at what this means in your environment
These are not options on a price list but part of the base configuration. What exactly they mean in your environment depends on your current system—let's look at it together.