Vultr Cloud GPUs Powered by NVIDIA A16: The Perfect Infrastructure for Virtual Desktops, Transcoding, and More
Feb 09, 2023

Today we’re pleased to introduce Vultr Cloud support for the NVIDIA A16 GPU, enabling an unprecedented user experience for virtual desktop infrastructure (VDI).

Vultr is the first cloud computing provider to offer the NVIDIA A16 GPU, and for us, this is the next step in our breakthrough work to democratize access to accelerated computing that we began last May. As we’ve done for the NVIDIA A100 Tensor Core GPU and the NVIDIA A40 GPU, we’re utilizing innovative virtualization to offer fractions of the NVIDIA A16 at unrivaled, affordable prices.

The end result is a game-changing product that puts high-performance, low-latency Windows and Linux virtual desktops at your fingertips, accessible from anywhere, at any time.

Remote, virtual desktops are finally really great

If it’s been a while since you last tried a virtual desktop, you should know that the technology has come a long way. And if you’ve never tried one before, there’s no better time than today!

VDI is not new, though the technology historically had a bit of a reputation for leaving users frustrated with latency and dropped frames. To a large extent, this is attributable to many past virtual desktops relying exclusively on traditional CPUs. Now, with a modern, purpose-built NVIDIA GPU attached, everything gets a whole lot better.

Our previously announced Cloud GPUs featuring the NVIDIA A40 make the perfect virtual workstations for the most demanding graphics-intensive applications, such as visual effects (VFX), rendering, and computer-aided design (CAD).

For less graphics-intensive applications – think developer IDEs like Visual Studio or Android Studio, or Microsoft PowerPoint, Excel, or Word – Cloud GPUs powered by the NVIDIA A16 pack more than enough punch. You might also try the A16 for some of the aforementioned graphics-intensive apps (and video games), and upgrade to an A40 if you find you need more power.

Vultr delivers a virtual desktop end-user experience that’s nearly indistinguishable from a native PC because we use the latest, greatest NVIDIA software and hardware together. Specifically, Vultr Cloud instances featuring the NVIDIA A16 utilize NVIDIA Virtual PC (vPC) software to accelerate desktop applications.

Considering that virtual desktop performance has come such a long way, it’s making more sense every day for businesses to embrace the technology. Remote work is now normalized for many professions, and IT departments will increasingly find that it is easier, more secure, and less expensive to manage virtual desktops than physical computers.

Affordable and versatile for transcoding, streaming, and machine learning inference

Through the wonders of virtualization, we’re able to slice up the NVIDIA A16 GPU into fractions. Here’s the full set of plans we’re launching today.

We’ll also note that, especially at these low price points, you might want to try using the NVIDIA A16 for use cases in addition to virtual desktops. The NVIDIA A16 features the highest number of video encoders and decoders with four on-chip hardware encoders (NVENC) and eight decoder (NVDEC) units in a single board. This makes the NVIDIA A16 ideal for all sorts of transcoding and streaming workloads. We’ve also found success using fractions of the A16 for machine learning inference.

NVIDIA A16, A40, and A100 GPUs also available as bare metal

The NVIDIA A16 is the latest addition to the growing collection of NVIDIA GPUs that Vultr makes available for use in both virtual machines and as bare metal. With bare metal, you’ll have direct access to all underlying physical hardware, including all the GPUs on the server, without any virtualization layer.

Get started today, or contact us to discuss your business needs

To provision a virtual machine or bare-metal server powered by the NVIDIA A16, A40, or A100 GPU, visit the Cloud GPU area within the Vultr control panel. The Vultr Cloud featuring the NVIDIA A16 is available today in New Jersey, London, Frankfurt, and Tokyo.

After provisioning a server with the NVIDIA A16 or A40, you can connect to it easily using a variety of clients, including Parsec, Microsoft Remote Desktop, and VNC. Then, you can access your Windows and Linux virtual desktops from any device, whether you’re using a Chromebook, a tablet, or even a smartphone.

We encourage businesses interested in large-scale deployments of Cloud GPUs – be it for VDI, AI, VFX, HPC, or any other use case – to contact our sales team to learn more. We've recently expanded our capacity for the NVIDIA A40 GPU, which is ideal for visual computing, and the NVIDIA A100 GPU, which is perfect for AI, machine learning, and HPC. The NVIDIA A40 through Vultr is currently available in Los Angeles, New Jersey, London, Frankfurt, Tokyo, Bangalore, Singapore, and Sydney. The NVIDIA A100 through Vultr is available in New Jersey, Silicon Valley, London, and Tokyo, and other locations are available upon request.

We also look forward to introducing support for the latest NVIDIA H100 Tensor Core GPUs. Please contact us here if you’d like to reserve these AI supercomputers.