Virtualization in 2026: Building a platform for VMs, containers, and AI

The past few years have been a period of significant change and challenges in the virtualization industry. Customers are forced to re-evaluate their virtualization platforms while adapting to the rapid rise of AI and infrastructure as well as supply chain pressures.

Even with all this turmoil, technology leaders and virtualization admins still need to manage the critical workloads, databases, applications, and virtual machines (VMs) within the budget constraints. If you’re wondering what's next for virtualization infrastructure, the answer is Red Hat OpenShift Virtualization. This strategic evolution serves as your platform for VMs, containers, and AI for the future. 

Over the past year, Red Hat has delivered the capabilities organizations rely on everyday—from multicluster management, intelligent load balancing, memory oversubscription, advanced networking, enterprise storage, operating system (OS) diversity, performant applications, and comprehensive security. We haven’t just built these features; we’ve helped organizations successfully move them into production. In fact more than 20 of these customers are presenting at Red Hat Summit 2026. This momentum is supported by a dramatically expanded partner ecosystem, offering integrated solutions across storage, backup, VDI, cloud, and database workloads.  

OpenShift Virtualization offers a strategic path forward that goes beyond simple licensing savings. It provides agentic intelligence, open source innovation, integrated DevOps and app platform, hybrid cloud across all major public clouds, automated upgrades with no downtime, and a platform that supports your entire application portfolio (e.g.,VMs, containers, and AI workloads). 

For those navigating industry uncertainty, OpenShift Virtualization is the proven platform for the future. Red Hat was recognized as a leader by 3 different vendors, including the Gartner Magic Quadrant for Container Management, the Gartner Magic Quadrant for Cloud-Native Application Platforms, and the Forrester Wave Multicloud Container Platform.

As a 2025 innovation award winner, OpenShift Virtualization delivers an ambitious roadmap without sacrificing the enterprise-grade stability you require. The future of virtualization is open, and we’re ready to help you escape vendor lock-in. 

Join us at Red Hat Summit 2026 to see our latest OpenShift Virtualization roadmap, live demos, and new partner announcements. 

What’s new in OpenShift Virtualization

Over the past year, we have introduced significant enhancements designed to simplify the administrator experience, automate workload mobility, and harden security.

User interface (UI) designed for virtual machine (VM) administrators

The new administrative interface abstracts Kubernetes complexity, providing a familiar workflow for managing VM lifecycles and infrastructure. 

• Virtualization focus view: A simplified UI that filters out container complexity, showing only VMs.
• Tree view navigation: Organizes and groups VMs by department, application, or custom hierarchies, reducing cognitive load for large VM inventories.
• Red Hat Advanced Cluster Management: Provides a single pane of glass for multicluster VM lifecycle operations across all your clusters. Users only see and manage VMs where they have explicit permissions. Red Hat Advanced Cluster Management provides observability and management of both VMs and containers from one platform.
• Guided networking: Network configurations reduce complexity.
• Ansible integration: Native inventory support in the Ansible Automation Platform UI allows you to transfer your existing automation skills directly to OpenShift VMs. You can manage VM lifecycle, Day 2 operations, and orchestration using Ansible playbooks and patterns.

Workload mobility and load balancing

Moves workloads where they perform best, with zero downtime and automated intelligence. 

• The platform enables CPU load-aware rebalancing to automatically distribute workloads, while providing advanced mobility through live migration with destination node selection and VM storage live migration. Cross-cluster VM live migrations eliminates maintenance windows.
• Live migration controls: Specifies destination nodes for migrations and automatic co-location of VMs with hard affinity rules to ensure application availability and performance requirements are met.
• Storage migration: Moves VMs between different storage classes without downtime, facilitating storage lifecycle management, new storage deployments, or decommissioning legacy infrastructure.
• Cross-cluster VM live migrations: Migrates VMs between OpenShift clusters with zero downtime across regions, clouds, and datacenters to eliminate maintenance windows.
• VM right-sizing feature: Reduces resource consumption while ensuring availability and reliability.

Advanced networking

• User-defined networks: Creates isolated, private networks within or across namespaces to give you the flexibility to connect your VMs together in the manner that best suits your organization’s needs.

Easing the migration journey

The migration toolkit for virtualization (MTV) helps bridge the gap from legacy hypervisors. 

• Network preservation during migration: Automatically handles the driver transition while maintaining network interface names and DHCP configuration for a “no-touch” migration.
• Comprehensive guest OS support: Migrates the entire VMware estate, including Windows, Linux, and legacy systems with clear documentation on what's supported and tested.
• VM template import: Accelerates post-migration deployments by bringing carefully crafted templates directly into OpenShift Virtualization.

Modern security fundamentals

We continue to set the standard for scale and density while maintaining a "security-first" posture.

• Modern Windows support: Full vTPM support with persistence allows for Windows 11 security features, such as BitLocker encryption, SecureBoot, and attestation.
• Hardening guide and CIS Benchmark: We have published the official CIS Benchmark® for Red Hat OpenShift Virtualization, accompanied by OpenShift Virtualization Hardening Guide. 

Performance focus on scale, density, and storage

• Our Performance Engineering team created numerous examples and tuning guides focused on performance and scalability. These include 10K VMs in a Cluster, Tuning & Scaling Guide, Out-of-the Box Scalable Database Performance , Dynamic VM CPU Load Aware Workload Rebalancing, Multi IOthread feature intro, I/O workload perf improvement examples, and Boost Database VM density with memory overcommit.

2026 roadmap and future innovations

Starting with our mid-2026 release, we’re prioritizing AI-empowered migrations, enhanced edge topologies, and deep hardware acceleration for AI workloads.

Ease of migration 

We are lowering the barrier to entry with self-service tools and smarter automation.

• OpenShift migration advisor: A no-cost, self-service tool that evaluates the migration-readiness of your workloads and generates technical reports to plan your migration with confidence.
• AI-enhanced MTV: The introduction of Red Hat Lightspeed (formerly known as Red Hat Insights) and dedicated MTV MCP AI agents assist with expanding hypervisor compatibility and improving storage and automation. Additional providers will support migration from most of the common hypervisors. 

Simplification

• Simplified deployment: The Assisted Installer simplifies the process of deploying connected OpenShift clusters with a guided, automated experience, enabling OpenShift Virtualization during cluster setup, with support extending to disconnected environments. With built-in pre-installation validations, it checks for hardware and system requirements (e.g., CPU virtualization flags, memory, storage, and consistent deployments), reducing the risk of errors, including disconnected cluster installations.
• Flexible VM lifecycle: Expanded template capabilities allow for OVA-style import/export, direct conversion of VMs to templates, and the use of InstanceTypes or full manifests to standardize complex configurations.
• Node management: Enhanced node management provides a unified view of the physical infrastructure, including host-level network configuration, local storage adaptors, and mounted PVs. Further details will include critical high-availability (HA) information and context-appropriate node performance data, as well as consolidated logs.

Data protection and disaster recovery

• Storage-agnostic incremental backups: QEMU Changed Block Tracking (CBT) for space and time-efficient backups regardless of storage provider. Multiple checkpoints provide backup and restore flexibility, with integration for Red Hat-certified data protection partners.
• Regional disaster recovery (DR): Vendor-native storage replication with Red Hat Advanced Cluster Management orchestration for failover and failback. Integration with CSI replication add-ons from storage vendors provides enterprise-grade DR without third-party software.
• Storage and DR: These enhancements enable non-disruptive storage operations and granular backups. Using native replication, the system ensures comprehensive DR with significantly reduced (Recovery Time Objectives) RTOs and (Recovery Point Objectives) RPOs.

Workload mobility and load balancing:

• Cluster migration: Streamlined cross-cluster migrations expanding network topologies via EVPN and network proxies. This framework additionally serves as the foundation for enabling cross-namespace live migrations.
• Storage offload: Reduced MTV migration times from hours to minutes, and expanded storage offload functionality to support live VM storage migrations.
• Rapid recovery capabilities: Reduced node failure recovery time from unplanned issues to less than 60 seconds. 

Hybrid cloud and edge expansion: 

• Multicloud availability: We are now generally available across Amazon Web Services (AWS), Google Cloud, Microsoft Azure, Oracle OCI, and IBM Cloud. We are currently expanding storage partnerships within these platforms and preparing for a new collaborative announcement with IBM Cloud.
• Edge computing: We are introducing a consistent two-node topology featuring native fencing mechanisms, extending the platform to small-scale edge environments like retail, logistics, and manufacturing. This architecture supports software-defined storage to enable fully hyperconverged solutions. 

Networking advancement

• Key networking pillars: Enhanced Primary UDN IP capabilities, complex cross-cluster Layer 2 connectivity with BGP and EVPN, EVPN fabric connectivity for VMs with OVN-K localnet, and accelerated data planes for telco and real-time workloads via OVS DPDK.
• Operational efficiency: Reduced network configuration friction with automatic detection of MTU misalignment for VMs.

Workload density

• Infrastructure optimization: Improved workload density, focusing on memory tiering as well as safely overcommitting resources like memory and CPU. Resource prioritization ensures critical workloads receive guaranteed CPU and memory slices during contention. Improved node use through scheduler tuning to allow for a higher density of VMs per node. 

Accelerated hardware 

• Strategic partnerships: Through collaboration with leading hardware providers, we provide out-of-the-box support for the latest AI accelerators. Starting in mid-2026, we will enable full GPU passthrough for NVIDIA ARM-based platforms, including Grace Hopper (GH200) and Grace Blackwell (GB200/GB300).
• Resource management: We ensure predictable performance via NUMA-aware GPU placement, multi-GPU passthrough with NVLink, and Kubernetes-native Dynamic Resource Allocation (DRA). Support for SR-IOV networking (ConnectX-7, 8, and 9) and a specialized kernel deliver near-bare-metal performance for AI workloads.
• GPU virtualization: We’re developing GPU-as-a-Service for Hosted Control Planes and adding support for SR-IOV vGPU. Upcoming features include live migration capabilities for vGPUs to ensure continuous uptime.

Resiliency and trust 

• Confidential computing: Introduction of confidential VMs using hardware-based memory encryption and security-hardened attestation, alongside deep Zero Trust Workload Identity Manager integration for hardware-rooted zero trust identities.
• Security and hardening: Implementation of hardened CIS profiles for virtualized pods and native Red Hat Advanced Cluster Security for Kubernetes vulnerability scanning for Linux-based guest VMs.
• Security integration: Zero Trust Workload Identity Manager for zero trust identity to enable a security-focused, hardware-rooted identity for VMs.
• Roadmap for security and hardening: Establishes a security-focused and compliant platform through 2 key initiatives. First, it introduces Hardened CIS Profile for virtualized pods to help ensure industry-standard compliance. Second, it integrates Red Hat Advanced Cluster Security for Kubernetes vulnerability reporting to provide native support for scanning and reporting on Linux-based guest VMs. 

A platform for your future

As you evaluate what’s next for your virtualization strategy, the path forward should be designed for where your business is going, not where it has been. OpenShift Virtualization delivers the capabilities you rely on today, while giving you the flexibility to evolve on your terms. With a strong foundation already in production and a roadmap focused on continued innovation, you can move forward with confidence. The future of virtualization is open, and Red Hat can help you achieve your goals.

Check out the resources below to get started with Red Hat OpenShift Virtualization. 

• Explore the documentation.
• Walk through these demos.
• Check out how customers are using our product.