<img height="1" width="1" style="display:none" src="https://www.facebook.com/tr?id=1678247062437069&amp;ev=PageView&amp;noscript=1">

AMD, Microsoft Azure, and Teradici Collaborate on Virtual Workstations with Flexible GPU Access

March 16, 2020
HP Teradici

HP Teradici is the inventor of the PCoIP remote display protocol and develops the Engineering Emmy-Award-winning HP Anyware (formerly Teradici CAS) to deliver the best virtual and remote desktop experience in the world.

This blog post was first published in November 2019 and updated March 2020 with new information.


Many of you have been following the progress of the Microsoft Azure NVv4 instance, which uses the AMD Radeon Infinity MI25 Accelerator and 2nd Gen AMD EPYC™ processors to offer full GPU access for power users and others whose applications perform better with a GPU, such as graphics intensive 3D modeling applications. We are pleased to confirm that this solution is now generally available from Microsoft and is fully supported by Teradici Cloud Access Software. More details about the solution are in the post below from our joint demonstration of it at Microsoft Ignite in November, and we now also have a reference architecture available to assist customers who are looking to deploy the new NVv4 instances with Cloud Access Software.

“The flexibility that Azure NVv4 with AMD-powered GPU partitioning provides for users to share and access GPU resources as needed is a valuable feature that we see will benefit many Teradici customers,” said Ziad Lammam, vice president of product management for Teradici. “We are excited to be working with Microsoft and AMD to enable more flexible, cost-effective GPU options for virtual desktop and virtual workstation use cases such as AEC.”

For more information, please review NVv4 documentation and pricing.


We’ve demonstrated many times how Teradici Cloud Access Software delivers the best virtual workstation experience for artists and other professionals working with graphics-intensive workloads hosted in a data center or cloud. And we’ve talked about how hosting virtual workstations in the cloud can be more cost and energy efficient by optimizing the use of the available cloud-based GPU resources and replacing desk-side workstations that are only used for part of the day. For the most part, however, we’ve limited our discussion of these use cases to roles that work extensively with graphic content for most or all of their time.


In reality, there are many other roles that can benefit from virtual workstations with access to cloud-based GPUs. Some of these are power users or knowledge workers that use Windows 10, which will offer a better user experience when graphics acceleration is available. In addition, as our regular computing tasks, our information sharing platforms, and our learning systems have become more graphics and visually oriented, many roles involve graphic-intensive work for at least some of the time, and those positions have been steadily increasing over the last several years.


Consider the marketing manager who a few years ago worked mainly with static images and printed words, but who now spends up to a quarter of her time editing video, for example. Or the executive assistant posting recorded video Q&A sessions that have replaced company-wide memos on corporate intranets. Or the HR intern working with Camtasia to produce an internal training video to onboard new hires. None of these are traditionally considered “graphics-intensive” roles and would not normally justify the extra cost of dedicated GPUs for their virtual machines. But all of them could benefit from using GPU resources when they need them, if they were available.


AMD and Microsoft have been collaborating on a solution for users like these. The new Microsoft Azure NVv4 instance with the AMD Radeon Instinct™ MI25 accelerator and 2nd Gen AMD EPYC™ Processors offers full GPU access to power users and others we commonly have accepted to be in need of a GPU for work like 3D modeling in computer assisted design roles, for example. But in addition to that, this solution enables virtual workstation users to share the GPU between as many as eight virtual machines, providing the flexibility of GPU access for users who don’t require a fully dedicated GPU all the time, along with the ability to scale up to full GPU processing power when it is needed.



If these users need GPU resources, even some of the time, then they also need Cloud Access Software to provide a high quality visualization experience for their graphics-intensive tasks. The combined Microsoft Azure, AMD Radeon and 2nd Gen AMD EPYC™ Processor solution above, which is now in technical preview, is also supported by Teradici Cloud Access Software. In providing these users with the resources they need in their most demanding use cases, even if those use cases only comprise a small portion of their overall computing needs, organizations benefit from the overall efficiency and security advantages that desktop virtualization provides without compromising the user experience.


Teradici will participate in an information session along with AMD, and Microsoft Azure team at Microsoft Ignite in Orlando this week - drop in to learn more or see us in the AMD booth on the show floor. We hope to see you there!

Interested in getting more information about Cloud Access Software? Provide us with a few details here and we’ll be in touch.

Interested in early access to the NVv4 series VMs? Fill out this form.

 Experience Cloud Access Software  Click here to schedule a 1-1 demo with one  of our PCoIP experts.


HP Teradici

HP Teradici is the inventor of the PCoIP remote display protocol and develops the Engineering Emmy-Award-winning HP Anyware (formerly Teradici CAS) to deliver the best virtual and remote desktop experience in the world.