Yesterday, we open-sourced ZSvirt. Today, we want to talk about why.
ZSvirt didn't start life as an open-source project. Its commercial counterpart, ZStack ZSphere, is a virtualization platform built by ZStack's engineering team over years of large-scale VMware replacement work — continuously developed, validated in real production environments, and deeply optimized.
The codebase now released as ZSvirt v1.0.0 has been running in demanding production environments for a long time. In these environments, the control plane manages more than 10,000 hosts, and a single VM can carry up to 768 vCPUs. For workloads this critical, makeshift maintenance is simply not an option.
That history matters to us, because virtualization is not a field where a demo alone proves long-term value.
A hypervisor's real worth is proven in critical moments and under pressure: when a disk fails, when you live-migrate during peak hours, when an audit deadline lands.
We've been through enough of these moments to believe that ZSvirt is ready for scrutiny from a wider community.
Why Now? Why Open Source?
Two years ago, this decision wouldn't have felt as urgent.
Since Broadcom acquired VMware, the way we talk to infrastructure administrators has changed. Teams running vSphere — and teams also running Hyper-V, Proxmox VE, or Nutanix — are re-evaluating their virtualization strategy, often under cost and budget pressure.
In these conversations, the questions we keep hearing are not about features. They are:
"Can we read the code?" "Can we audit it?" "If the vendor's strategy changes down the road, how do we protect ourselves?"
We could have answered these questions with promises. Instead, we chose to answer with a public repository. ZSvirt v1.0.0 is the first release to open the complete source code, build scripts, a signed installation ISO, and the accompanying documentation.
More importantly, in our view, "zero vendor lock-in" should live in the license and the code itself — not as a tagline in a product brochure. The virtualization layer carries a large share of an enterprise's business-critical workloads. If users can't read, audit, or build from the code themselves, a critical part of their infrastructure remains a dependency they don't fully control. Our belief is simple: infrastructure this important should not be subject to any single vendor's commercial decisions — including our own.
We understand that people may have questions about this decision, because we'd ask the same ones. An infrastructure open-source project launched by a commercial company naturally raises legitimate concerns: will the license change as the project grows? Is this an open-core play that keeps the real capabilities hidden?
So we want to state our commitments publicly.
A Transparent Business Model
We will clearly define the boundary between the community edition and the commercial offering.
What stays open, always:
- The virtualization platform itself, including compute, storage, and network virtualization;
- The full API surface;
- ZMigrate, our free and unlimited migration tool;
- The Terraform Provider and Go, Python, and Java SDKs.
What's commercial:
- Enterprise-grade support services from ZStack;
- Capabilities exclusive to the enterprise edition.
We will maintain public documentation that spells out which capabilities belong to the community edition and which to the commercial edition — and we will not move already-open features behind a paywall. That commitment applies to future releases as well.
On the license: ZSvirt will continue to be released under GPL 3.0. The license will not change. We want that to be clear from day one.
Where We Fit
We don't believe ZSvirt is the only right choice for every organization, and we're not here to dismiss approaches the industry has widely adopted. Organizations have different infrastructure needs. ZSvirt is built for teams and individuals who want enterprise-grade capability, openness, and control at the same time.
There are several valid paths in virtualization, each serving different operational needs:
- Public cloud managed services: the provider runs operations; you pay per workload.
- Hyperconverged infrastructure platforms: integrated hardware and software from a single vendor.
- A self-built KVM stack: for teams with deep Linux expertise willing to build their own toolchain.
- Kubernetes-native virtualization: for organizations all-in on Kubernetes, managing infrastructure cloud-natively.
- Lightweight single-node virtualization: for small deployments and edge scenarios that don't need a full cluster control plane.
Many enterprise teams, though, have a different set of needs. They want to keep the familiar operating model of a traditional virtualization platform while gaining the openness, flexibility, and control that come with an open foundation. These teams typically manage infrastructure around familiar concepts — clusters, hosts, VMs, networks, and datastores — and need production-grade capabilities. At the same time, they want to avoid hardware and platform lock-in, don't want Kubernetes as a prerequisite for managing VMs, and want a fully open API.
For these teams, ZSvirt offers:
- Scale: a single control plane managing 10,000+ hosts;
- Openness: 2,000+ asynchronous RESTful/OpenAPI endpoints, with Go, Python, and Java SDKs;
- Security: native Secure Boot, vTPM 2.0, VM encryption, and built-in key management;
- Networking: distributed switches and port groups, plus SR-IOV-accelerated networking with live migration support;
- Migration: ZMigrate, an agentless migration tool running at the virtualization layer, cutting over in as little as 5 minutes;
- Enterprise capabilities: HA, DRS, snapshots, backup, alerting, and more.
If your organization is looking for that balance, ZSvirt is built for you. If another approach fits your needs better, we respect that — and encourage you to adopt the platform that best suits your environment.
Our Community
For us, open-sourcing ZSvirt is not a one-time launch event. It's the start of a long-term collaboration.
Concretely:
- A public roadmap, updated quarterly, so users and contributors can see where the platform is heading;
- Fully public contribution guidelines — contributions of every kind are welcome, from docs and testing to code;
- A predictable community process: technical discussions happen on GitHub Discussions; bugs and documentation issues are tracked in GitHub Issues, with the engineering team responsible for each module participating directly. We respond to issues on business days.
We look forward to the community shaping where ZSvirt goes next — open collaboration builds stronger software.
Try ZSvirt
Download the ISO — published with checksums and cryptographic signatures — and explore ZSvirt in your own environment:
- Source code: https://github.com/zsvirt/zsvirt
- Downloads: https://zsvirt.io/download
- Documentation: https://zsvirt.io/docs
Star or watch the ZSvirt repository to follow new releases, development progress, and technical discussions.
Join the Community
- Issue reports: GitHub Issues
- Community discussions: GitHub Discussions
-
Social media:
- YouTube: https://youtube.com/@ZSvirt
- LinkedIn: https://www.linkedin.com/in/zsvirt-community/
- X/Twitter: https://x.com/ZSvirt
- Reddit: https://www.reddit.com/r/ZSvirt/
- Email: community@zsvirt.io
