Virtual GPU computing power
Publish: 2021-04-26 03:22:03
1. Just pretend it doesn't matter. Remote desktop certainly has a great influence when running, sunflower will consume GPU computing power to transmit data
2. Computing power is a concept put forward by NVIDIA when it released CUDA (Compute Unified Device Architecture, a programming language for GPU, similar to C programming for CPU). Because the graphics card itself is a floating-point computing chip, it can be used as a computing card, so the graphics card has computing power. Different graphics cards have different computing power. In order to show the difference, NVIDIA put forward the corresponding version of computing power x.x on the procts of different periods. Computing power 1.0 appeared on early graphics cards, such as the original 8800 ultra and many 8000 Series cards, as well as Tesla C / D / s870s cards. Cuda1.0 was released corresponding to these graphics cards. Today, computing power 1.0 has been eliminated from the market. Then there was computing power 1.1, which appeared on many 9000 Series graphics cards. Computing power 1.2 appears together with GT200 Series graphics card, while computing power 1.3 is proposed when upgrading from GT200 to GT200 A / b revision. In the future, there will be computing power 2.0, 2.1, 3.0 and other versions. The latest released version is computing power 6.1, which is supported by the latest Pascal architecture graphics card. At the same time, CUDA version is also updated to cuda8.0
ordinary users do not need to care about the computing power of the graphics card, only GPU programmers care about this problem when they write CUDA programs to develop GPU computing. As long as you know the model of your computer's graphics card, you can find the corresponding computing power https://developer.nvidia.com/cuda-gpus .
ordinary users do not need to care about the computing power of the graphics card, only GPU programmers care about this problem when they write CUDA programs to develop GPU computing. As long as you know the model of your computer's graphics card, you can find the corresponding computing power https://developer.nvidia.com/cuda-gpus .
3. Different manufacturers have different ways. Generally, they need to install specific drivers and start specific services. In addition, it depends on whether the graphics card supports vgpu
4. GPU virtualization means that VDI can go into areas that it couldn't before - a big news for users who are still on the sidelines with desktop delivery technology
the virtual image processing unit (GPU) card can help share the graphics processing load of the server and improve the performance of applications, so the virtual desktop infrastructure (VDI) can now be applied to more types of users
users who use 3D or CAD, as well as video intensive and game applications, without this technology to rece processor load, their VDI performance will not be guaranteed
"without this technology, CAD will not be able to run in VDI," he said
before the appearance of virtual GPU technology, VDI was mainly used to deal with some daily work. Before many users put forward higher requirements for GPU processing performance, desktop virtualization provided by Citrix and VMware can meet the needs of 60% to 70% of users. Said Justin Boitano, marketing director of NVIDIA, a leading provider of virtualization GPU technology
"but they can be a blow to users with graphics processing needs, and they can't fully meet those needs," he said
in order to meet the increasing demand of users for performance, desktop virtualization providers have seized this opportunity to support NVIDIA's grid technology. Citrix has recently added hardware GPU sharing function for xend desktop 7, and VMware has introced virtual private graphics accelerator feature in view 5.3, both of which are based on grid. In addition, Amazon Web Services launched its second-generation elastic cloud computing proct in November, which also supports grid and enables GPU acceleration in the cloud<
the practical application of virtual GPU technology
the IT Department of Florida Atlantic University started using VDI four years ago to try to provide remote access to 3-D applications for students and professors in graphics and game programming courses, which accounted for 30% of the total number of students. They install ATI graphics cards and Teradici hardware accelerators on physical workstations - but that's not enough, says mahesh neelakanta, director of technical services at tau. This kind of deployment will take up a lot of physical resources, and only allow one-to-one connection between users and machines
"GPU virtualization is changing all this."
the University installed NVIDIA K1 and K2 boards earlier this year, which provides higher flexibility and integration features, allowing it to connect 8 to 12 users on each shared GPU board at the same time
"by using VDI to deploy images, we have been able to rece our IT requirements. And provide virtual desktop with 3D acceleration function for all students. "
with virtual GPU technology promoting desktop virtualization to more users, the utilization rate of VDI will be higher in the next few years. However, companies must ensure that GPU virtualization can help them. For example, if you transmit video, you may only need to optimize the remote display protocol,
the virtual image processing unit (GPU) card can help share the graphics processing load of the server and improve the performance of applications, so the virtual desktop infrastructure (VDI) can now be applied to more types of users
users who use 3D or CAD, as well as video intensive and game applications, without this technology to rece processor load, their VDI performance will not be guaranteed
"without this technology, CAD will not be able to run in VDI," he said
before the appearance of virtual GPU technology, VDI was mainly used to deal with some daily work. Before many users put forward higher requirements for GPU processing performance, desktop virtualization provided by Citrix and VMware can meet the needs of 60% to 70% of users. Said Justin Boitano, marketing director of NVIDIA, a leading provider of virtualization GPU technology
"but they can be a blow to users with graphics processing needs, and they can't fully meet those needs," he said
in order to meet the increasing demand of users for performance, desktop virtualization providers have seized this opportunity to support NVIDIA's grid technology. Citrix has recently added hardware GPU sharing function for xend desktop 7, and VMware has introced virtual private graphics accelerator feature in view 5.3, both of which are based on grid. In addition, Amazon Web Services launched its second-generation elastic cloud computing proct in November, which also supports grid and enables GPU acceleration in the cloud<
the practical application of virtual GPU technology
the IT Department of Florida Atlantic University started using VDI four years ago to try to provide remote access to 3-D applications for students and professors in graphics and game programming courses, which accounted for 30% of the total number of students. They install ATI graphics cards and Teradici hardware accelerators on physical workstations - but that's not enough, says mahesh neelakanta, director of technical services at tau. This kind of deployment will take up a lot of physical resources, and only allow one-to-one connection between users and machines
"GPU virtualization is changing all this."
the University installed NVIDIA K1 and K2 boards earlier this year, which provides higher flexibility and integration features, allowing it to connect 8 to 12 users on each shared GPU board at the same time
"by using VDI to deploy images, we have been able to rece our IT requirements. And provide virtual desktop with 3D acceleration function for all students. "
with virtual GPU technology promoting desktop virtualization to more users, the utilization rate of VDI will be higher in the next few years. However, companies must ensure that GPU virtualization can help them. For example, if you transmit video, you may only need to optimize the remote display protocol,
5. Comparison of virtual GPU features between Citrix and VMware
at present, the two most popular enterprise VDI procts are based on NVIDIA's grid technology. What are the differences between them
Citrix XenDesktop grid vgpu
some experts believe that Citrix's GPU virtualization technology is ahead of other competitors because it is supported by NVIDIA at the lowest level. In Citrix XenDesktop 7.1, Citrix provides native support for vgpu technology by using NVIDIA grid technology, including pass through and GPU sharing. However, it should be noted that NVIDIA's board does not support too many service
processors, and the grid vgpu feature is only available on XenServer 6.2 with service pack 1 installed
after you deploy the vgpu for the XenDesktop virtual machine (VM), your main concern is how to allocate the GPU. You can allocate physical GPUs to virtual GPUs in a one-to-one way, which can rece the load on each physical GPU and achieve high performance of applications; Or you can share a physical GPU into multiple virtual GPUs, which can provide high-density GPUs, but
the performance will be reced. Again, it all depends on the needs of the application and the user
the grid card can provide four different types of vgpu. Each one varies according to the number of shared users that need to be supported and the amount of memory required. Grid card also provides GPU pass through mode, which is the only GPU type that xenapp can use
VMware vsga and vdga
VMware technologies are also based on NVIDIA, including different GPU virtualization solutions: soft 3D, vsga and vdga
soft 3D uses software to simulate GPU, which is suitable for limited 3D image processing needs
because it divides the GPU into multiple parts and allocates memory to each virtual machine, vsga can support high user density. It can also provide the function of fail over. This feature uses the same driver as soft
3D and can automatically switch to soft 3D mode when needed. You can choose whether to use only hardware GPU or allow automatic switch to soft 3D
if you want to achieve the best performance, you can use VMware vdga, which assigns each GPU to only one virtual desktop. The highest executive standards of DirectX and OpenGL are also used
an interesting thing to note is that Citrix VDI can run in multiple hypervisors, so if Citrix XenDesktop runs in VMware esxi, it can also use vsga and vdga
new features of GPU
NVIDIA's vgpu only supported Citrix
XenServer for a long time, but NVIDIA announced that it would expand its support for VMware vSphere
hypervisor by 2015. For applications that run in the cloud and need a lot of graphics processing power, VMware and NVIDIA also cooperate to add grid support to horizon DAAS (desktop as a service) platform
NVIDIA has continuously improved its GPU procts, including increasing user density, developing Citrix receiver software client for GPU accelerator running NVIDIA Tegra processor, and increasing support for windows 8.1 and windows server 2012 R2 by updating vgpu.
at present, the two most popular enterprise VDI procts are based on NVIDIA's grid technology. What are the differences between them
Citrix XenDesktop grid vgpu
some experts believe that Citrix's GPU virtualization technology is ahead of other competitors because it is supported by NVIDIA at the lowest level. In Citrix XenDesktop 7.1, Citrix provides native support for vgpu technology by using NVIDIA grid technology, including pass through and GPU sharing. However, it should be noted that NVIDIA's board does not support too many service
processors, and the grid vgpu feature is only available on XenServer 6.2 with service pack 1 installed
after you deploy the vgpu for the XenDesktop virtual machine (VM), your main concern is how to allocate the GPU. You can allocate physical GPUs to virtual GPUs in a one-to-one way, which can rece the load on each physical GPU and achieve high performance of applications; Or you can share a physical GPU into multiple virtual GPUs, which can provide high-density GPUs, but
the performance will be reced. Again, it all depends on the needs of the application and the user
the grid card can provide four different types of vgpu. Each one varies according to the number of shared users that need to be supported and the amount of memory required. Grid card also provides GPU pass through mode, which is the only GPU type that xenapp can use
VMware vsga and vdga
VMware technologies are also based on NVIDIA, including different GPU virtualization solutions: soft 3D, vsga and vdga
soft 3D uses software to simulate GPU, which is suitable for limited 3D image processing needs
because it divides the GPU into multiple parts and allocates memory to each virtual machine, vsga can support high user density. It can also provide the function of fail over. This feature uses the same driver as soft
3D and can automatically switch to soft 3D mode when needed. You can choose whether to use only hardware GPU or allow automatic switch to soft 3D
if you want to achieve the best performance, you can use VMware vdga, which assigns each GPU to only one virtual desktop. The highest executive standards of DirectX and OpenGL are also used
an interesting thing to note is that Citrix VDI can run in multiple hypervisors, so if Citrix XenDesktop runs in VMware esxi, it can also use vsga and vdga
new features of GPU
NVIDIA's vgpu only supported Citrix
XenServer for a long time, but NVIDIA announced that it would expand its support for VMware vSphere
hypervisor by 2015. For applications that run in the cloud and need a lot of graphics processing power, VMware and NVIDIA also cooperate to add grid support to horizon DAAS (desktop as a service) platform
NVIDIA has continuously improved its GPU procts, including increasing user density, developing Citrix receiver software client for GPU accelerator running NVIDIA Tegra processor, and increasing support for windows 8.1 and windows server 2012 R2 by updating vgpu.
6. This is a new concept. The weakness of previous virtualization is graphics processing, especially in instries with higher requirements for graphics cards such as 3D. There are technical problems in realizing desktop virtualization. GPU virtualization needs graphics workstation group processing, the client does not install anything, the cost of course is much higher than server, desktop, application virtualization.
7.
After the data center is equipped with NVIDIA vgx platform, employees can access a real cloud PC using any device such as thin client, notebook, tablet, smartphone, etc., regardless of the operating system used by such devices. Employees can easily use all kinds of applications previously only used on office PCs
NVIDIA vgx enables knowledge workers to access GPU accelerated desktops similar to traditional local PCs for the first time. The platform's management options and ultra-low latency remote display capabilities bring great convenience to users who use 3D design and simulation tools, which previous virtual desktops were not competent for
after integrating the vgx platform into the enterprise network, the byod problem that plagues the enterprise IT department will be solved, and employees will no longer have to bring their own computing tools to the workplace. Vgx platform provides a remote desktop, which enables users to use such tools in the same way as they operate their own desktop terminals. At the same time, it helps to rece overall it overhead, improve data security, and minimize data center complexity
8. Mininet tutorial Abstract: here gives the mininet in U environment test preparation in advance: host Ubuntu + virtual machine KVM (the official image is vmdk VMware format, the tutorial gives vmdk to KVM image qcow2 installation) operating system: after the completion of the installation
9. Support models
nvida's graphics cards supporting multi os technology include:
al Quadro 6000, 5000, 4000, 2000
FX 5800
FX 4800
FX 3800
nvida's graphics cards supporting multi os technology include:
al Quadro 6000, 5000, 4000, 2000
FX 5800
FX 4800
FX 3800
10. There are few graphics cards supported by GPU virtualization, and this GPU virtualization, I feel, is a proct of VMware
because you (MS) have it, so I also want to have it. The actual value is not big.
because you (MS) have it, so I also want to have it. The actual value is not big.
Hot content