NVIDIA GRID with New Tesla M10 Meets Growing Demand for Graphics-Accelerated Virtualization
OREANDA-NEWS. August 29, 2016. Increasingly businesses are seeing the benefits of enterprise-wide virtualization for all employees. To meet this growing demand, NVIDIA GRID with the Tesla M10 was designed to help IT departments deliver virtualized desktops to all of their employees at an affordable cost
Shipping in volume today, the Tesla M10 provides the industry’s highest user density on a single GPU — 64 users per board, or up to 128 users per server with two boards. Knowledge workers using modern business applications — such as Outlook, Office 2016, web browsers, Adobe Photoshop and the Windows 10 operating system — will enjoy the best virtual user experience available.
Used in combination with the GRID August 2016 release including our new NVIDIA GRID Monitoring, the Tesla M10 gives businesses the most extensive vGPU control and visibility at both the host and guest levels. Together, they provide businesses the actionable insights necessary to fully leverage their investments in graphics-accelerated virtualization.
Today, there are more than 23 certified systems supporting the Tesla M10 from leading vendors, such as Dell, Nutanix and Supermicro.
Early testing has shown the Tesla M10 significantly improves performance across applications, including streaming video, webGL and graphical web apps, compared to CPU-only systems. In addition, it reduces CPU usage up to 18%, freeing up resources for other tasks.(1)
“Our initial testing of the Tesla M10 shows us that we can fit 2x the number of users in one of our GPU servers with no sacrifice in performance,” said Matt Wahl, infrastructure engineer at Autodesk.
NVIDIA GRID with the Tesla M10 will be on display at NVIDIA booth 955 at VMworld, Aug. 29-Sept. 1. The Dell PowerEdge R730 with the Tesla M10 will be in Dell’s booth 1431. To learn more about moving to 100% graphics virtualization in your enterprise, attend our VMworld talks. Follow all the show news at #VMworld.
(1) Based on internal subjective testing using the NVIDIA GRID vPC 512MB (the smallest available) profile on the Tesla M10.