Best Android Solutions Get Tips to Install, Rooting, And Trick About Android Phone

GPU virtualization

Nvida Gpu Ai

Nvidia has announced an upcoming video platform for graphics rendering – – based on cloud computing and GPU virtualization – at the GPU Technology Conference that was held on May 14-17 in San Jose, CA. This platform is called the Nvidia VGX and offers the potential for mobile devices like the iPad and iPhone to have the same full graphical power of the most powerful desktop workstations.

According to Nvidia s press release announcing this platform, the GPUs can be as powerful as Nvidia wants them to be and are housed in enterprise data centers. Users are able to connect remotely to this hardware and can access its power without having to purchase a very expensive product and install it in a powerful enough system to run it. Widgets The term bottlenecking refers to when a computer cannot take advantage of a GPU due to not having enough hardware horsepower. This eliminates this problem entirely because any device connected remotely to this virtual server can access the same power as if they were sitting in front of it. The other thing that this form of cloud connectivity and virtualization allows for users accessing the GPU is scalability.

A company can purchase a license to use such a service from Nvidia for 15 employees, for example, at a cost of $15 per employee, instead of spending thousands of dollars on dedicated GPUs for each employee (not to mention the systems that will house them). This has been carried out with other computing infrastructure services, also known as laaS (infrastructure as a service), but this is the first time a GPU company like Nvidia has planned to offer such a service.

The press release mentions 100 users being able to connect to the virtual server and GPU platform simultaneously.

According to a report from Wired, Nvidia has actually teamed up with Citrix to make this platform a possibility. Citrix will be providing Nvidia with XenDesktop and XenServer hypervisor desktop virtualization.

Now, before delving into what potential this form of GPU virtualization has in store for our favorite products like the iPads or Macbook Airs, let s discuss the history and definition of the GPU.


What Is Gpu Virtualization

GPUs stand for graphical – processing units, otherwise known as video cards. They were invented in the early 90s as a way for PC gaming to be able to match and even exceed consoles of the time like the original Playstation and the Nintendo 64. Previously, gamers had to rely on software rendering and this was really taxing on the CPU.

For the CPU to be freed up and allowed to concentrate on other tasks within the software, external hardware needed to be purchased and inserted into the machines. This hardware is the GPU or video card that enables complex modeling to be created like shapes with many polygons, shadows, bitmaps, shader effects, anti – aliasing and more.

The early 3D video cards included names such as 3DFX s Voodoo series. However, the early video cards only had 2D capabilities rather than allowing for any 3D rendering.

The term GPU was not coined until Nvidia came up with the GeForce 256 video card that left most other previous 3D solutions in the dark.

GPUs evolved from just hardware dedicated to providing 3D graphics support and targeting mostly gamers to a wide range of users and markets.


Apple has not really been using Nvidia GPUs as of late in any hardware. Macs rely on AMD GPUs in the Radeon lineup, while the iDevices rely on an ARM – based SOC (system – on – chip) with a PowerVR GPU. Nvidia is nowhere to be seen in Apple products. However, the company has been producing Tegra SOCs (also Arm – based), which include Nvidia GPUs housed within them. Because of this conflict of interest, it may take some time for Apple and Nvidia to support the VGX GPU platform to run on iPads.

However, expect AMD or competitor to come up with a solution shortly and Apple to take full advantage of this. If you have been active in the PC gaming scene, you may remember how Nvidia had the 3D Vision platform exclusively for quite some time before AMD came up with competing solutions in that realm with the AMD HD3D.

However, Nvidia does offer some GPU solutions for Macs when it comes to the heavy workstation productivity market. An example of such a video card is the Nvidia Quadro 4000 for Mac. It is geared for uses such as 3D visual effects post – production using software like Maya, and it currently costs $1,199.95 from the Apple Store.

The upcoming Nvidia CGX platform will also be aimed at enterprise users and scalability at first, so Apple may benefit in the end. It would be quite an overkill to use such a GPU on a mobile device, however, especially without the software support to back up its use. If Nvidia does allow such scalability, as they claim they may with tablets mentioned in the press release, maybe the software support will follow, although doubtful outside the gaming and consumer markets due to other constraints like lacking multi – touch precision (when compared to a mouse) and display size.


A company called OnLive has demonstrated the kind of power and potential cloud computing and virtualization of hardware have on the consumer market. The company released a gaming platform called OnLive that relies on entire server machines and not just GPUs to be scaled across devices from users across the Internet. The games are then streamed online to the users7 devices of choice.

The company later released a Microsoft Office solution for the iPad called OnLive Desktop. This also showed the potential that remote server connectivity has on user productivity. There is no need to purchase full applications or have a supporting operating system. A user on the iPad can take advantage of the Office suite without a single install of the actual software. The iPad simply opens up a window to a virtual machine sitting at a data center somewhere else in the world.

However, one aspect that many users complain about when trying to play games using OnLive is latency and bandwidth. For users who do not have the fastest connections, they can experience some latency or lag issues between the games response to user input. Bandwidth caps are also an issue that service providers limit and OnLive requires a lot of bandwidth consumption per use.

Nvidia addresses at least one of the issues in the VGX press release. The company states that this upcoming platform will offer flexibility and high performance for end users accessing the power of the remote GPUs.

Gaming was not mentioned as Nvidia s aim with the platform in the press release, so expect Nvidia to target the high-end video post – production market – at least at first. However, this hasn t stopped many bloggers from speculating on where the technology will lead.

According to Wired, “While the VGX cloud platform is intended for enterprise applications, Nvidia plans to use it for cloud gaming.” Take that as you will, but gaming is Nvidia s forte and money maker so this is a very high possibility.


Nvidia is making a smart move by offering high-end power from GPUs costing over a grand to the fingertips of the masses. This will open up new markets and allow companies to scale production more efficiently. It will also expand Nvidia s products to more users and allow more users to understand this market of high-end video processing. There are still some questions that remain, however. How will tablet or smartphone users even take advantage of such high-end video processing and what software would they utilize with it? Will Nvidia target the consumer market and offer possibly consumer GPUs aimed at gamers? It is an interesting time in computing, that s for sure. Details regarding the release date of this technology are still unclear.

Want to buy good and cheap shoes? Then visit the Birkenstock chaussures and choose what suits you. It really is a quality shoe that will serve you for a very long time. So hurry to buy and be happy!