Showing results for 
Search instead for 
Did you mean: 

Quality of Service in Virtualized Graphics Deployments Using AMD MxGPU Technology

0 0 1,289


[Originally posted on 02/28/18]

AMD MxGPU graphics virtualization technology enables leading Quality of Service (QoS) with performance that is up to nine times more consistent than the competition1.

Reliable, predictable infrastructure is paramount in an enterprise environment. IT managers demand robust QoS to meet the needs of a cloud-connected workforce. AMD Multi-User GPU (MxGPU) technology was designed to meet this demand by delivering a predictable and consistent end-user experience.


A predictable graphics experience for digital workspaces yields value across an organization. IT managers get fewer calls probing unexpected slowdowns. End users know how their workflow will behave and the Program Management Office can practice better business planning through predictable resource allocation.

As more independent software vendors adopt GPU acceleration to provide a richer experience for end-users, GPU-bound QoS will continue to be top-of-mind for enterprise data centers. AMD MxGPU technology offers an elegant solution, enabling world class QoS with AMD FirePro™ S7100-series graphics accelerators.

AMD MxGPU technology, the world’s first hardware-based virtualized GPU solution, is built on industry standard SR-IOV (Single-Root I/O Virtualization) technology and allows up to 16 virtualized users per physical GPU to work remotely.

To learn more about the QoS benefits AMD MxGPU technology brings to your data center, check out the infographic below.


Visit our page on AMD MxGPU technology and download our QoS whitepaper.

Michael DeNeffe, AMD’s Director of Cloud Graphics and has extensive experience in network computing, and client device technologies. Links to third-party sites and references to third-party trademarks are provided for convenience and illustrative purposes only. Unless explicitly stated, AMD is not responsible for the contents of such links, and no third-party endorsement of AMD or any of its products is implied. Use of third-party names or marks is for informational purposes only and no endorsement of or by AMD is intended or implied.


1 - With 16 virtual machines, AMD FirePro™ S7150 x2 delivers up to 9x lower standard deviation in SPECapc® for SOLIDWORKS® 2015 than NVIDIA® Tesla M60, derived using Graphics Composite Scores from SPECapc® for SOLIDWORKS® 2015. Testing conducted by AMD Performance Labs as of August 1st, 2017 under following host and virtual machine configuration: Host: VMware® ESXi™ 6.5.0 with host driver Radeon™ Pro VMware vSphere® Driver – Revision 1.02/NVIDIA® VMware® ESXi™ 6.5 Host Driver 367.106. Virtual Machines: Intel(R) Xeon(R) E5-2698 v4 @2.20 GHz, 4 Cores, 4 Logical Processors, 8GB Physical Memory, 120 GB Storage, Windows 7 Enterprise 64-bit SP1 Build 7601 with Radeon™ Pro Software 17.30 driver/NVIDIA® GRID 370.12 guest graphics driver. Benchmark: SPECapc® for SOLIDWORKS® 2015 simultaneously on 16 virtual machines. NVIDIA® GRID vGPU configuration: Each VM has 1 vGPU assigned from NVIDIA® GRID M60-0Q profile (16 vGPU on a physical GPU). AMD MxGPU configuration: Each VM has 1 vGPU assigned from 16 VFs configuration (16 vGPU on a physical GPU). AMD Graphics Composite Scores, Enumerated Per Virtual Machine Instance: 0.32, 0.32, 0.32, 0.33, 0.32, 0.32, 0.32, 0.33, 0.32, 0.32, 0.32, 0.32, 0.33, 0.32, 0.32, 0.32. NVIDIA® Graphics Composite Scores Results, Enumerated Per Virtual Machine Instance: 0.41, 0.37, 0.31, 0.32, 0.37, 0.32, 0.32, 0.31, 0.37, 0.33, 0.41, 0.29, 0.33, 0.39, 0.33, 0.37. AMD FirePro™ S7150 x2 standard deviation of the Graphic Composite scores: 0.00390. NVIDIA® Tesla M60 standard deviation of the Graphic Composite scores: 0.03618. Performance Differential: 0.03618/0.00390 = ~9.27x higher standard deviation on NVIDIA® Tesla M60. Additional information about the SPEC benchmarks can be found at PC and server manufacturers may vary configurations, yielding different results. Performance may vary based on use of latest drivers. RPW-177

© 2018 Advanced Micro Devices, Inc. All rights reserved. AMD, the AMD Arrow logo, Radeon, and combinations thereof are trademarks of Advanced Micro Devices, Inc. SPEC® and the benchmark SPECapc® are registered trademarks of Standard Performance Evaluation Corporation. Other product names used in this publication are for identification purposes only and may be trademarks of their respective companies.

About the Author
Michael DeNeffe is AMD’s Director of Cloud Graphics and has over 25 years of experience in network computing, and client device technologies. He has also held senior management positions at NVIDIA, Teradici, Wyse, Transmeta, and NEC. Over the last ten years, Michael has focused on the emerging and dynamic desktop virtualization market. Prior to is position at AMD, he led the worldwide product marketing effort for NVIDIA GRID technology. Prior to that he was Vice President of Marketing and Business Development for Teradici Corporation and was instrumental in bringing to market the innovative PC-over-IP technology and protocol. And before that, he led the product management and marketing teams for Wyse Technology. There, he launched numerous thin computing products and earned a U. S. patent for the innovative design which is still in the market today. Michael has both undergraduate and master’s degrees from Santa Clara University.