UNSOLVED
N
NumaoMasayuki
9 Posts
0
15192
November 2nd, 2010 03:00
GPU virtualization on M610X
Hi, I am using M610X with TESLA C1060 GPU. I want to use the GPU from guest OS.
As a host OS, I use Hyper-V on Windows Server 2008 R2, I also installed TESLA driver and the GPU is visible from the host OS. But from the guest OS, which is also Windows Server 2008 R2, the device manager cannot find the GPU, and TESLA installer failed to install.
Is there anyway to use the GPU from guest OS?
As a host OS, I use Hyper-V on Windows Server 2008 R2, I also installed TESLA driver and the GPU is visible from the host OS. But from the guest OS, which is also Windows Server 2008 R2, the device manager cannot find the GPU, and TESLA installer failed to install.
Is there anyway to use the GPU from guest OS?
- This is what I got back from the NVIDIA SME:
***
Windows server 2008 R2 Hyper-V does not support direct assignment of GPUs into guest VMs.
Citrix XenServer is capable of doing this, and we’re working with Citrix to fully qualify Tesla/Quadro GPUs used on XenServer.
Windows Server 2008 R2 “SP1”, when released, will introduce Remote FX, which will work with M2070Q, not M1060 or M2050. The GPU is still not visible in the VM directly, but Windows Hyper-V does include a Microsoft RemoteFX Graphics Device.
****
Here's a URL with info on Windows Server 2008 R2 SP1: http://news.softpedia.com/news/Windows-7-SP1-RTM-in-Q1-2011-148260.shtml NumaoMasayuki
9 Posts
409
0
Posted November 7th, 2010 19:00
Thank you for the information.
Then, I will try XenServer. Is there any URL for GPU virtualization on XenServer?
Kong Yang
180 Posts
409
0
Posted November 3rd, 2010 19:00
Checking with the SMEs for you.