Unified cloud infrastructure platform that brings together compute, storage, networking, security, identity, operations, and lifecycle management to deploy, operate, and scale the ZStack portfolio with a consistent architecture.
How to install ZCF?
1
2
3
1
Install ZStack Cloud
REQUIREDCOREv5.5.22
A next-generation IaaS software that adheres to the 4S standards (Simple, Strong, Smart, Scalable) and provides powerful private cloud features and addon services.
A unified installation and deployment tool for ZCF. It guides basic configurations and component installation to help quickly set up a ZCF environment.
Enterprise virtualization foundation for VM lifecycle management, high availability, live migration, distributed networking, and storage integration on standard servers.
How to install ZVF?
1
2
1
ZStack ZSphere
REQUIREDv5.0.2
Core ZVF package for deploying production virtualization environments with centralized VM operations.
VMware AlternativeLive Migrationx86 and ARM Support
Consolidates compute, storage, and networking into standard servers — with unified management, linear scaling, and enterprise-grade reliability.
One platform replaces three separate systems.
Replace separate compute servers, SAN arrays, and network switches with one integrated stack. Deploy on commodity servers, scale by adding nodes with zero downtime, and manage everything through a single control plane.
3-node minimum — deploy a full HCI cluster from just three servers, scale linearly
VMware migration — built-in V2V migration, UI-driven, no scripting required
RPO = 0 — distributed storage extension clusters for critical system disaster recovery
High performance — NUMA binding, large-page memory, OVS-DPDK and SR-IOV acceleration
One-click upgrades — cross-version upgrades with zero service interruption
AIOS
ZStack AIOS
Private AI infrastructure with GPU scheduling, model training, inference, and AI operations. Your data never leaves your environment.
Run enterprise AI on your own infrastructure.
Three integrated layers — compute power, model services, and operations — give you a complete private AI platform. Precise GPU scheduling down to 1% granularity, 95% passthrough performance, and full lifecycle model management from training to production inference.
