Linux Virtualization — KVM, QEMU, libvirt, virtio & microVMs
Parent: DevOps, Infrastructure & Observability · researched 2026-06-01T05:45:08.680Z· 22 sources · 10 concepts · skill devops-infra
DevOps / infrastructure / observability family ROUTER. Split into focused sub-hubs — route to: devops-linux-internals (kernel, boot, memory/NUMA, storage/filesystems, virtualization, io_uring, cgroups
devops-infra
- DevOps / infrastructure / observability family ROUTER. Split into focused sub-hubs - route to: devops-linux-internals (kernel, boot, memory/NUMA, storage/filesystems, virtualization, io_uring, cgroups/namespaces, sandboxing, immutable Linux, privilege); devops-linux-admin (sysadmin, systemd, package management, shell scripting, host networking diagnostics); devops-containers-cicd (Docker, Kubernetes, CI/CD pipelines, Terraform/Kafka IaC, git workflows, library packaging); devops-observability (Node/OTel observability, Pino logging, Sentry, eBPF, Linux perf tracing). Pick the sub-hub matching the task. [source]
- This hub routes to on-demand reference files under references/. See each spoke for depth. [source]
Children
- KVM in-kernel hypervisor (/dev/kvm ioctl API — KVM_CREATE_VM/VCPU/RUN, Intel VT-x/VMX root vs non-root, AMD-V/SVM, EPT/NPT two-dimensional paging, VM exits, posted interrupts/APICv, one-thread-per-vCPU, nested virt) (frontier)
- QEMU userspace VMM & device model (accelerators kvm/tcg/hvf, machine types q35/pc/virt/microvm versioned, -machine/-cpu/-device/-drive/-netdev model, OVMF/UEFI firmware, qcow2 vs raw, io_uring block backend) (frontier)
- virtio paravirtualized device framework (virtqueues/vrings split vs packed, feature negotiation, transports virtio-pci/-mmio/-ccw, device family net/blk/scsi/fs/gpu/balloon/vsock/rng) (frontier)
- virtio data-plane acceleration ladder (QEMU-emulated → vhost-net/vhost-scsi kernel → vhost-user DPDK/SPDK → vDPA hardware offload) (frontier)
- libvirt management layer (domain XML + virsh, the modular daemons virtqemud/virtnetworkd/virtnodedevd/virtstoraged/virtsecretd, virtproxyd, default NAT network + storage pools) (frontier)
- microVMs (Firecracker 5-device model + jailer + ~125ms boot/<5MiB, Cloud Hypervisor, rust-vmm crates, QEMU microvm machine type, Kata wrapping) (frontier)
- Device assignment (VFIO/vfio-pci, IOMMU groups + VT-d/AMD-Vi, SR-IOV PF/VF, GPU passthrough, mediated devices mdev) (frontier)
- Live migration (pre-copy iterative + stop-and-copy, post-copy demand fault, CPU-model/machine-type compatibility, VFIO device migration) (frontier)
- Confidential VMs (AMD SEV/SEV-ES/SEV-SNP via PSP, Intel TDX/SEAM Trust Domains, guest memory encryption + integrity + remote attestation) (frontier)
- VM performance tuning (host-passthrough vs host-model, 1:1 CPU pinning, NUMA affinity, 2M/1G hugepages, multiqueue virtio-net, vhost, iothreads, balloon/KSM overcommit) (frontier)
Frontier under this node: Confidential VMs (AMD SEV/SEV-ES/SEV-SNP via PSP, Intel TDX/SEAM Trust Domains, guest memory encryption + integrity + remote attestation), Device assignment (VFIO/vfio-pci, IOMMU groups + VT-d/AMD-Vi, SR-IOV PF/VF, GPU passthrough, mediated devices mdev), KVM in-kernel hypervisor (/dev/kvm ioctl API — KVM_CREATE_VM/VCPU/RUN, Intel VT-x/VMX root vs non-root, AMD-V/SVM, EPT/NPT two-dimensional paging, VM exits, posted interrupts/APICv, one-thread-per-vCPU, nested virt), Live migration (pre-copy iterative + stop-and-copy, post-copy demand fault, CPU-model/machine-type compatibility, VFIO device migration), QEMU userspace VMM & device model (accelerators kvm/tcg/hvf, machine types q35/pc/virt/microvm versioned, -machine/-cpu/-device/-drive/-netdev model, OVMF/UEFI firmware, qcow2 vs raw, io_uring block backend), VM performance tuning (host-passthrough vs host-model, 1:1 CPU pinning, NUMA affinity, 2M/1G hugepages, multiqueue virtio-net, vhost, iothreads, balloon/KSM overcommit), libvirt management layer (domain XML + virsh, the modular daemons virtqemud/virtnetworkd/virtnodedevd/virtstoraged/virtsecretd, virtproxyd, default NAT network + storage pools), microVMs (Firecracker 5-device model + jailer + ~125ms boot/<5MiB, Cloud Hypervisor, rust-vmm crates, QEMU microvm machine type, Kata wrapping), virtio data-plane acceleration ladder (QEMU-emulated → vhost-net/vhost-scsi kernel → vhost-user DPDK/SPDK → vDPA hardware offload), virtio paravirtualized device framework (virtqueues/vrings split vs packed, feature negotiation, transports virtio-pci/-mmio/-ccw, device family net/blk/scsi/fs/gpu/balloon/vsock/rng)