GPU virtualization refers to technologies that allow the use of a
GPU to
accelerate
In mechanics, acceleration is the rate of change of the velocity of an object with respect to time. Accelerations are vector quantities (in that they have magnitude and direction). The orientation of an object's acceleration is given by the ...
graphics
Graphics () are visual images or designs on some surface, such as a wall, canvas, screen, paper, or stone, to inform, illustrate, or entertain. In contemporary usage, it includes a pictorial representation of data, as in design and manufacture, ...
or
GPGPU
General-purpose computing on graphics processing units (GPGPU, or less often GPGP) is the use of a graphics processing unit (GPU), which typically handles computation only for computer graphics, to perform computation in applications traditiona ...
applications running on a
virtual machine
In computing, a virtual machine (VM) is the virtualization/ emulation of a computer system. Virtual machines are based on computer architectures and provide functionality of a physical computer. Their implementations may involve specialized har ...
. GPU virtualization is used in various applications such as
desktop virtualization
Desktop virtualization is a software technology that separates the desktop environment and associated application software from the physical client device that is used to access it.
Desktop virtualization can be used in conjunction with applicati ...
,
cloud gaming
Cloud gaming, sometimes called gaming on demand or gaming-as-a-service or game streaming, is a type of online gaming that runs video games on remote servers and streams them directly to a user's device, or more colloquially, playing a game rem ...
and
computational science
Computational science, also known as scientific computing or scientific computation (SC), is a field in mathematics that uses advanced computing capabilities to understand and solve complex problems. It is an area of science that spans many disc ...
(e.g.
hydrodynamics simulations).
GPU virtualization implementations generally involve one or more of the following techniques: device emulation, API remoting, fixed pass-through and mediated pass-through. Each technique presents different trade-offs regarding virtual machine to GPU
consolidation ratio,
graphics
Graphics () are visual images or designs on some surface, such as a wall, canvas, screen, paper, or stone, to inform, illustrate, or entertain. In contemporary usage, it includes a pictorial representation of data, as in design and manufacture, ...
acceleration
In mechanics, acceleration is the rate of change of the velocity of an object with respect to time. Accelerations are vector quantities (in that they have magnitude and direction). The orientation of an object's acceleration is given by ...
, rendering
fidelity
Fidelity is the quality of faithfulness or loyalty. Its original meaning regarded duty in a broader sense than the related concept of ''fealty''. Both derive from the Latin word ''fidēlis'', meaning "faithful or loyal". In the City of London ...
and
feature support,
portability
Portability may refer to:
*Portability (social security), the portability of social security benefits
* Porting, the ability of a computer program to be ported from one system to another in computer science
** Software portability, the portability ...
to different hardware,
isolation between virtual machines, and support for suspending/resuming and
live migration.
API remoting
In
API remoting or API forwarding, calls to graphical APIs from guest applications are forwarded to the host by
remote procedure call
In distributed computing, a remote procedure call (RPC) is when a computer program causes a procedure (subroutine) to execute in a different address space (commonly on another computer on a shared network), which is coded as if it were a normal (lo ...
, and the host then executes graphical commands from multiple guests using the host's GPU as a single user.
It may be considered a form of
paravirtualization
In computing, paravirtualization or para-virtualization is a virtualization technique that presents a software interface to the virtual machines which is similar, yet not identical, to the underlying hardware–software interface.
The intent o ...
when combined with device emulation.
This technique allows sharing GPU resources between multiple guests and the host when the GPU does not support hardware-assisted virtualization. It is conceptually simple to implement, but it has several disadvantages:
* In pure API remoting, there is little isolation between virtual machines when accessing graphical APIs; isolation can be improved using paravirtualization
* Performance ranges from 86% to as low as 12% of native performance in applications that issue a large number of drawing calls per
frame
* A large number of API
entry points must be forwarded, and partial implementation of entry points may decrease fidelity
* Applications on guest machines may be limited to few available APIs
Hypervisors usually use
shared memory
In computer science, shared memory is memory that may be simultaneously accessed by multiple programs with an intent to provide communication among them or avoid redundant copies. Shared memory is an efficient means of passing data between progr ...
between guest and host to maximize performance and minimize latency. Using a
network interface Network interface may refer to:
* Network interface controller, a computer hardware component that connects a computer to a computer network
* Network interface device, a device that serves as the demarcation point between a telephone carrier's loc ...
instead (a common approach in
distributed rendering Parallel rendering (or distributed rendering) is the application of parallel programming to the computational domain of computer graphics. Rendering graphics can require massive computational resources for complex scenes that arise in scientific v ...
), third-party software can add support for specific APIs (e.g.
rCUDA
rCUDA, which stands for Remote CUDA, is a type of middleware software framework for remote GPU virtualization. Fully compatible with the CUDA application programming interface (API), it allows the allocation of one or more CUDA-enabled GPUs to a ...
for
CUDA
CUDA (or Compute Unified Device Architecture) is a parallel computing platform and application programming interface (API) that allows software to use certain types of graphics processing units (GPUs) for general purpose processing, an approach ...
) or add support for typical APIs (e.g.
VMGL for
OpenGL
OpenGL (Open Graphics Library) is a cross-language, cross-platform application programming interface (API) for rendering 2D and 3D vector graphics. The API is typically used to interact with a graphics processing unit (GPU), to achieve ha ...
) when it is not supported by the hypervisor's software package, although
network delay and
serialization
In computing, serialization (or serialisation) is the process of translating a data structure or object state into a format that can be stored (e.g. files in secondary storage devices, data buffers in primary storage devices) or transmitted (e ...
overhead may outweigh the benefits.
Fixed pass-through
In fixed pass-through or GPU pass-through (a special case of
PCI
PCI may refer to:
Business and economics
* Payment card industry, businesses associated with debit, credit, and other payment cards
** Payment Card Industry Data Security Standard, a set of security requirements for credit card processors
* Prov ...
pass-through), a GPU is accessed directly by a single virtual machine exclusively and permanently. This technique achieves 96100% of native performance
and high fidelity,
but the acceleration provided by the GPU cannot be shared between multiple virtual machines. As such, it has the lowest
consolidation ratio and the highest cost, as each graphics-accelerated virtual machine requires an additional physical GPU.
The following software technologies implement fixed pass-through:
*
VMware
VMware, Inc. is an American cloud computing and virtualization technology company with headquarters in Palo Alto, California. VMware was the first commercially successful company to virtualize the x86 architecture.
VMware's desktop software ru ...
Virtual Dedicated Graphics Acceleration (vDGA)
*
Parallels Workstation Extreme
*
Hyper-V
Microsoft Hyper-V, codenamed Viridian, and briefly known before its release as Windows Server Virtualization, is a native hypervisor; it can create virtual machines on x86-64 systems running Windows. Starting with Windows 8, Hyper-V superseded Wi ...
Discrete Device Assignment (DDA)
*
Citrix
Citrix Systems, Inc. is an American multinational cloud computing and virtualization technology company that provides server, application and desktop virtualization, networking, software as a service (SaaS), and cloud computing technologies. C ...
XenServer GPU pass-through
*
Xen and
QEMU
QEMU is a free and open-source emulator (Quick EMUlator). It emulates the machine's processor through dynamic binary translation and provides a set of different hardware and device models for the machine, enabling it to run a variety of g ...
/
KVM with
Intel GVT-d
VirtualBox
Oracle VM VirtualBox (formerly Sun VirtualBox, Sun xVM VirtualBox and Innotek VirtualBox) is a type-2 hypervisor for x86 virtualization developed by Oracle Corporation.
VirtualBox was originally created by Innotek GmbH, which was acquired by ...
removed support for PCI pass-through in version 6.1.0.
QEMU/KVM
For certain GPU models, Nvidia and AMD video card drivers attempt to detect the GPU is being accessed by a virtual machine and disable some or all GPU features.
NVIDIA has recently changed virtualization rules for consumer GPUs by disabling the check in GeForce Game Ready driver 465.xx and later.
For NVIDIA, various architectures of desktop and laptop consumer GPUs can be passed through in various ways. For desktop graphics cards, passthrough can be done via the KVM using either the legacy or UEFI BIOS configuration via SeaBIOS and OVMF, respectively.
NVIDIA
Desktops
For desktops, most graphics cards can be passed through, although for graphics cards with the Pascal architecture or older, the VBIOS of the graphics card must be passed through in the virtual machine if the GPU is used to boot the host.
Laptops
For laptops, the NVIDIA driver checks for the presence of a battery via ACPI, and without a battery, an error will be returned. To avoid this, an acpitable created from text converted into Base64 is required to spoof a battery and bypass the check.
= Pascal and earlier
=
For the laptop graphics cards that are Pascal and older, passthrough varies widely on the configuration of the graphics card. For laptops that do not have NVIDIA Optimus, such as the MXM variants, passthrough can be achieved through traditional methods. For laptops that have NVIDIA Optimus on as well as rendering through the CPU's integrated graphics framebuffer as opposed to its own, the passthrough is more complicated, requiring a remote rendering display or service, the use of Intel GVT-g, as well as integrating the VBIOS into the boot configuration due to the VBIOS being present in the laptop's system BIOS as opposed to the GPU itself. For laptops that have a GPU with NVIDIA Optimus and have a dedicated framebuffer, the configurations may vary. If NVIDIA Optimus can be switched off, then passthrough is possible through traditional means. However, if Optimus is the only configuration, then it is most likely that the VBIOS is present in the laptop's system BIOS, requiring the same steps as the laptop rendering only on the integrated graphics framebuffer, but an external monitor is also possible.
Mediated pass-through
In mediated device pass-through or full GPU virtualization, the GPU hardware provides
contexts with
virtual memory ranges for each guest through
IOMMU and the hypervisor sends graphical commands from guests directly to the GPU. This technique is a form of
hardware-assisted virtualization
In computing, hardware-assisted virtualization is a platform virtualization approach that enables efficient full virtualization using help from hardware capabilities, primarily from the host processors. A full virtualization is used to emulate a c ...
and achieves near-native performance and high fidelity. If the hardware exposes contexts as full logical devices, then guests can use any API. Otherwise, APIs and drivers must manage the additional complexity of GPU contexts. As a disadvantage, there may be little isolation between virtual machines when accessing GPU resources.
The following software and hardware technologies implement mediated pass-through:
*
VMware
VMware, Inc. is an American cloud computing and virtualization technology company with headquarters in Palo Alto, California. VMware was the first commercially successful company to virtualize the x86 architecture.
VMware's desktop software ru ...
Virtual Shared Pass-Through Graphics Acceleration with Nvidia vGPU
or AMD MxGPU
*
Citrix
Citrix Systems, Inc. is an American multinational cloud computing and virtualization technology company that provides server, application and desktop virtualization, networking, software as a service (SaaS), and cloud computing technologies. C ...
XenServer shared GPU with Nvidia vGPU, AMD MxGPU or Intel GVT-g
*
Xen and
KVM with
Intel GVT-g
* Thincast Workstation - Virtual 3D feature (Direct X 12 & Vulkan 3D API)
While API remoting is generally available for current and older GPUs, mediated pass-through requires hardware support available only on specific devices.
Device emulation
GPU architectures are very complex and change quickly, and their internal details are often kept secret. It is generally not feasible to fully virtualize new generations of GPUs, only older and simpler generations. For example,
PCem, a specialized emulator of the
IBM PC
The IBM Personal Computer (model 5150, commonly known as the IBM PC) is the first microcomputer released in the IBM PC model line and the basis for the IBM PC compatible de facto standard. Released on August 12, 1981, it was created by a tea ...
architecture, can emulate a
S3 ViRGE/DX graphics device, which supports
Direct3D
Direct3D is a graphics application programming interface (API) for Microsoft Windows. Part of DirectX, Direct3D is used to render three-dimensional graphics in applications where performance is important, such as games. Direct3D uses hardware ...
3, and a
3dfx Voodoo2, which supports
Glide, among others.
When using a
VGA or an
SVGA
Super VGA (SVGA) is a broad term that covers a wide range of computer display standards that extended IBM's VGA specification.
When used as shorthand for a resolution, as VGA and XGA often are, SVGA refers to a resolution of 800×600.
History
...
virtual display adapter,
the guest may not have 3D graphics acceleration, providing only minimal functionality to allow access to the machine via a graphics terminal. The emulated device may expose only basic 2D graphics modes to guests. The virtual machine manager may also provide common API implementations using
software rendering
Software rendering is the process of generating an image from a model by means of computer software. In the context of computer graphics rendering, software rendering refers to a rendering process that is not dependent upon graphics hardware ...
to enable 3D graphics applications on the guest, albeit at speeds that may be low as 3% of hardware-accelerated native performance.
The following software technologies implement graphics APIs using software rendering:
*
VMware
VMware, Inc. is an American cloud computing and virtualization technology company with headquarters in Palo Alto, California. VMware was the first commercially successful company to virtualize the x86 architecture.
VMware's desktop software ru ...
SVGA 3D software renderer
*
VirtualBox
Oracle VM VirtualBox (formerly Sun VirtualBox, Sun xVM VirtualBox and Innotek VirtualBox) is a type-2 hypervisor for x86 virtualization developed by Oracle Corporation.
VirtualBox was originally created by Innotek GmbH, which was acquired by ...
VMSVGA graphics controller
*
Citrix
Citrix Systems, Inc. is an American multinational cloud computing and virtualization technology company that provides server, application and desktop virtualization, networking, software as a service (SaaS), and cloud computing technologies. C ...
XenServer OpenGL Software Accelerator
*
Windows Advanced Rasterization Platform
Windows Advanced Rasterization Platform (WARP) is a software rasterizer and a component of DirectX graphics runtime in Windows 7 and later. It is available for Windows Vista and Windows Server 2008 through platform update for Windows Vista.
WARP ...
*
Core OpenGL software renderer
*
Mesa
A mesa is an isolated, flat-topped elevation, ridge or hill, which is bounded from all sides by steep escarpments and stands distinctly above a surrounding plain. Mesas characteristically consist of flat-lying soft sedimentary rocks capped by a ...
software renderer
See also
*
Comparison of platform virtualization software
Notes
References
{{reflist
Hardware virtualization