Our commitment
ZCP is built on open-source infrastructure: software with public source code, community governance, and no single vendor controlling your access to it.
We chose boring, inspectable systems with long production histories: Apache CloudStack for IaaS orchestration, KVM for virtualization, Ceph for storage, Kubernetes for containers, and Prometheus, Grafana, Alertmanager, and Loki for observability.
Why this foundation is credible
| Layer | What we use | External proof point |
|---|---|---|
| IaaS orchestration | Apache CloudStack | Apache lists CloudStack users across public cloud providers, product vendors, private cloud operators, research, and systems integrators. |
| Virtualization | KVM | KVM is open-source virtualization technology for Linux, allowing Linux to operate as a hypervisor. |
| Storage | Ceph | The Linux Foundation launched the Ceph Foundation with 30+ members, including Canonical, DigitalOcean, Intel, OVH, Red Hat, SUSE, Western Digital, and ZTE. |
| Kubernetes | Upstream Kubernetes | CNCF reported 82% of container users running Kubernetes in production in its latest annual cloud native survey. |
| Monitoring | Prometheus | Prometheus was the second CNCF project to graduate after Kubernetes, a maturity signal for governance and adoption. |
| Dashboards | Grafana | Grafana Labs reports more than 35 million users and 7,000+ customers. |
These proof points do not make ZSoftly famous by association. They show the foundation under ZCP is not experimental. The product risk is our operational execution, not whether the core infrastructure software is mature enough for production.
What open source means for you
No lock-in. Your virtual machines use a standard hypervisor format. Your object storage uses the S3 API. Your Kubernetes clusters run standard upstream manifests. If you leave ZCP, your workloads and data leave with you using tools you already own.
No proprietary agents. Nothing vendor-specific runs inside your VMs. You get root access to a clean machine. We have no persistent footprint inside your workload.
No black boxes. Every component of our infrastructure has publicly auditable source code. Review how it handles your data, check its CVE history, and verify its behaviour independently.
Proven at scale. The platforms we use are the same category of systems trusted by service providers, telecoms, research organizations, and enterprises globally. They are governed by established open-source foundations and communities, not a single vendor roadmap.
Your team already knows it. We deliberately chose tooling your engineers encounter in the broader industry. No proprietary training, no vendor-specific certifications required to operate on ZCP.
What we deliver
| Capability | What you get |
|---|---|
| Compute | Standard virtual machines with root access, hourly billing, NVMe storage |
| Block storage | Persistent volumes, online expansion, on-demand snapshots |
| Object storage | S3-compatible API, works with any S3 tool without modification |
| Kubernetes | Managed control plane, standard upstream Kubernetes, no forked distribution |
| Networking | L2, Isolated, and VPC networks; floating IPs, load balancers, firewall rules |
| Observability | Metrics, dashboards, alerting, log aggregation |
Sources
- Apache CloudStack users
- Apache CloudStack overview
- Red Hat: What is KVM?
- Linux Foundation: Ceph Foundation launch
- CNCF Annual Cloud Native Survey
- CNCF: Prometheus graduation
- Grafana Labs company milestone
Regions
| Region | Status |
|---|---|
| YUL-1 | Live. Primary region for all production workloads. |
| YOW-1 | Live. Dev/test and private cloud demos only. Non-redundant power and connectivity. Not recommended for production workloads. |
| YVR | Planned. |
| BUF | Planned. |
| LAX | Planned. |
| LHR | Planned. |
| AMS | Planned. |
YUL-1 and YOW-1 are live today, with five more regions on the roadmap across North America and Europe. All infrastructure is owned and operated by ZSoftly engineers end to end. There is no hyperscaler underneath ZCP and no reseller layer in the contract.
Frequently asked questions
Do you use AWS, Azure, or another hyperscaler underneath?
No. ZCP runs on hardware we own and operate. There is no hyperscaler, reseller arrangement, or cloud-on-cloud layer underneath.
Does Terraform work with ZCP resources?
Yes. Our infrastructure supports API-based automation and is compatible with standard infrastructure-as-code tooling. Contact us to confirm compatibility with your specific pipeline before migrating production workloads.
Is your object storage S3-compatible?
Yes. Any tool working with S3 (AWS CLI, boto3, rclone, s5cmd) works with ZCP object storage without modification beyond the endpoint URL.
What happens to my data if I cancel?
You retain full ownership of your data and export it using standard tools at any time. We do not hold data hostage or charge egress fees for account closure exports.
Is ZSoftly SOC 2 or ISO 27001 certified?
ZCP production runs in a Montreal Tier III data center with a SOC 2 Type II attestation and ISO 27001 certification. This describes the facility controls, not a claim that ZSoftly itself holds those certifications. If a company-level certification is a procurement requirement, contact us to discuss your timeline and the controls we provide today.
Why won’t you name your specific platforms?
We operate infrastructure, not a technology showcase. What matters to you is workload reliability, selected-region data placement, and no lock-in. The specific software behind delivery is an operational detail, not a customer concern.