1 Virtualization technology #
Virtualization is a technology that provides a way for a machine (Host) to run another operating system (guest virtual machines) on top of the host operating system.
1.1 Overview #
SUSE Linux Enterprise Server includes the latest open source virtualization technologies, Xen and KVM. With these hypervisors, SUSE Linux Enterprise Server can be used to provision, de-provision, install, monitor and manage multiple virtual machines (VM Guests) on a single physical system (for more information see Hypervisor). SUSE Linux Enterprise Server can create virtual machines running both modified, highly tuned, paravirtualized operating systems and fully virtualized unmodified operating systems.
The primary component of the operating system that enables virtualization is a hypervisor (or virtual machine manager), which is a layer of software that runs directly on server hardware. It controls platform resources, sharing them among multiple VM Guests and their operating systems by presenting virtualized hardware interfaces to each VM Guest.
SUSE Linux Enterprise is an enterprise-class Linux server operating system that offers two types of hypervisors: Xen and KVM.
SUSE Linux Enterprise Server with Xen or KVM acts as a virtualization host server (VHS) that supports VM Guests with its own guest operating systems. The SUSE VM Guest architecture consists of a hypervisor and management components that constitute the VHS, which runs many application-hosting VM Guests.
In Xen, the management components run in a privileged VM Guest often called Dom0. In KVM, where the Linux kernel acts as the hypervisor, the management components run directly on the VHS.
1.2 Virtualization benefits #
Virtualization brings a lot of advantages while providing the same service as a hardware server.
First, it reduces the cost of your infrastructure. Servers are mainly used to provide a service to a customer, and a virtualized operating system can provide the same service, with:
Less hardware: you can run several operating systems on a single host, therefore all hardware maintenance is reduced.
Less power/cooling: less hardware means you do not need to invest more in electric power, backup power, and cooling if you need more service.
Save space: your data center space is saved because you do not need more hardware servers (less servers than service running).
Less management: using a VM Guest simplifies the administration of your infrastructure.
Agility and productivity: Virtualization provides migration capabilities, live migration and snapshots. These features reduce downtime, and bring an easy way to move your service from one place to another without any service interruption.
1.3 Virtualization modes #
Guest operating systems are hosted on virtual machines in either full virtualization (FV) mode or paravirtual (PV) mode. Each virtualization mode has advantages and disadvantages.
Full virtualization mode lets virtual machines run unmodified operating systems, such as Windows* Server 2003. It can use either Binary Translation or hardware-assisted virtualization technology, such as AMD* Virtualization or Intel* Virtualization Technology. Using hardware assistance allows for better performance on processors that support it.
Certain guest operating systems hosted in full virtualization mode can be configured to use drivers from the SUSE Virtual Machine Drivers Pack (VMDP) instead of drivers originating from the operating system. Running virtual machine drivers improves performance dramatically on guest operating systems, such as Windows Server 2003. For more information, see Appendix A, Virtual machine drivers.
To be able to run under paravirtual mode, guest operating systems normally need to be modified for the virtualization environment. However, operating systems running in paravirtual mode have better performance than those running under full virtualization.
Operating systems currently modified to run in paravirtual mode are called paravirtualized operating systems and include SUSE Linux Enterprise Server.
1.4 I/O virtualization #
VM Guests not only share CPU and memory resources of the host system, but also the I/O subsystem. Because software I/O virtualization techniques deliver less performance than bare metal, hardware solutions that deliver almost “native” performance have been developed recently. SUSE Linux Enterprise Server supports the following I/O virtualization techniques:
- Full virtualization
Fully Virtualized (FV) drivers emulate widely supported real devices, which can be used with an existing driver in the VM Guest. The guest is also called Hardware Virtual Machine (HVM). Since the physical device on the VM Host Server may differ from the emulated one, the hypervisor needs to process all I/O operations before handing them over to the physical device. Therefore all I/O operations need to traverse two software layers, a process that not only significantly impacts I/O performance, but also consumes CPU time.
- Paravirtualization
Paravirtualization (PV) allows direct communication between the hypervisor and the VM Guest. With less overhead involved, performance is much better than with full virtualization. However, paravirtualization requires either the guest operating system to be modified to support the paravirtualization API or paravirtualized drivers. See Section 7.4.1, “Availability of paravirtualized drivers” for a list of guest operating systems supporting paravirtualization.
- PVHVM
This type of virtualization enhances HVM (see Full virtualization) with paravirtualized (PV) drivers, and PV interrupt and timer handling.
- VFIO
VFIO stands for Virtual Function I/O and is a new user-level driver framework for Linux. It replaces the traditional KVM PCI Pass-Through device assignment. The VFIO driver exposes direct device access to user space in a secure memory (IOMMU) protected environment. With VFIO, a VM Guest can directly access hardware devices on the VM Host Server (pass-through), avoiding performance issues caused by emulation in performance critical paths. This method does not allow to share devices—each device can only be assigned to a single VM Guest. VFIO needs to be supported by the VM Host Server CPU, chipset and the BIOS/EFI.
Compared to the legacy KVM PCI device assignment, VFIO has the following advantages:
Resource access is compatible with UEFI Secure Boot.
Device is isolated and its memory access protected.
Offers a user space device driver with more flexible device ownership model.
Is independent of KVM technology, and not bound to x86 architecture only.
In SUSE Linux Enterprise Server the USB and PCI pass-through methods of device assignment are considered deprecated and are superseded by the VFIO model.
- SR-IOV
The latest I/O virtualization technique, Single Root I/O Virtualization SR-IOV combines the benefits of the aforementioned techniques—performance and the ability to share a device with several VM Guests. SR-IOV requires special I/O devices, that are capable of replicating resources so they appear as multiple separate devices. Each such “pseudo” device can be directly used by a single guest. However, for network cards for example the number of concurrent queues that can be used is limited, potentially reducing performance for the VM Guest compared to paravirtualized drivers. On the VM Host Server, SR-IOV must be supported by the I/O device, the CPU and chipset, the BIOS/EFI and the hypervisor—for setup instructions see Section 14.12, “Assigning a host PCI device to a VM Guest”.
To be able to use the VFIO and SR-IOV features, the VM Host Server needs to fulfill the following requirements:
IOMMU needs to be enabled in the BIOS/EFI.
For Intel CPUs, the kernel parameter
intel_iommu=onneeds to be provided on the kernel command line. For more information, see https://github.com/torvalds/linux/blob/master/Documentation/admin-guide/kernel-parameters.txt#L1951.The VFIO infrastructure needs to be available. This can be achieved by loading the kernel module
vfio_pci. For more information, see Abschnitt 19.6.4, „Laden der Kernelmodule“.