<div dir="ltr"><div>If I am not wrong those are 2 GPUs</div><div><br></div><div>"tesla-v100:1" means 1 GPU <br></div><div><br></div><div>So e.g. a flavor with "pci_passthrough:alias": "tesla-v100:2"} will be used to create an instance with 2 GPUs</div><div><div><br></div></div><div>Cheers, Massimo</div></div><br><div class="gmail_quote"><div dir="ltr" class="gmail_attr">On Tue, Jan 18, 2022 at 11:35 PM Satish Patel <<a href="mailto:satish.txt@gmail.com">satish.txt@gmail.com</a>> wrote:<br></div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">Thank you for the information. I have a quick question.<br>
<br>
[root@gpu01 ~]# lspci | grep -i nv<br>
5e:00.0 3D controller: NVIDIA Corporation GV100GL [Tesla V100S PCIe<br>
32GB] (rev a1)<br>
d8:00.0 3D controller: NVIDIA Corporation GV100GL [Tesla V100S PCIe<br>
32GB] (rev a1)<br>
<br>
In the above output showing two cards does that mean they are physical<br>
two or just BUS representation.<br>
<br>
Also i have the following entry in openstack flavor, does :1 means<br>
first GPU card?<br>
<br>
{"gpu-node": "true", "pci_passthrough:alias": "tesla-v100:1"}<br>
<br>
<br>
<br>
<br>
<br>
<br>
On Tue, Jan 18, 2022 at 5:55 AM António Paulo <<a href="mailto:antonio.paulo@cern.ch" target="_blank">antonio.paulo@cern.ch</a>> wrote:<br>
><br>
> Hey Satish, Gustavo,<br>
><br>
> Just to clarify a bit on point 3, you will have to buy a vGPU license<br>
> per card and this gives you access to all the downloads you need through<br>
> NVIDIA's web dashboard -- both the host and guest drivers as well as the<br>
> license server setup files.<br>
><br>
> Cheers,<br>
> António<br>
><br>
> On 18/01/22 02:46, Satish Patel wrote:<br>
> > Thank you so much! This is what I was looking for. It is very odd that<br>
> > we buy a pricey card but then we have to buy a license to make those<br>
> > features available.<br>
> ><br>
> > On Mon, Jan 17, 2022 at 2:07 PM Gustavo Faganello Santos<br>
> > <<a href="mailto:gustavofaganello.santos@windriver.com" target="_blank">gustavofaganello.santos@windriver.com</a>> wrote:<br>
> >><br>
> >> Hello, Satish.<br>
> >><br>
> >> I've been working with vGPU lately and I believe I can answer your<br>
> >> questions:<br>
> >><br>
> >> 1. As you pointed out in question #2, the pci-passthrough will allocate<br>
> >> the entire physical GPU to one single guest VM, while vGPU allows you to<br>
> >> spawn from 1 to several VMs using the same physical GPU, depending on<br>
> >> the vGPU type you choose (check NVIDIA docs to see which vGPU types the<br>
> >> Tesla V100 supports and their properties);<br>
> >> 2. Correct;<br>
> >> 3. To use vGPU, you need vGPU drivers installed on the platform where<br>
> >> your deployment of OpenStack is running AND in the VMs, so there are two<br>
> >> drivers to be installed in order to use the feature. I believe both of<br>
> >> them have to be purchased from NVIDIA in order to be used, and you would<br>
> >> also have to deploy an NVIDIA licensing server in order to validate the<br>
> >> licenses of the drivers running in the VMs.<br>
> >> 4. You can see what the instructions are for each of these scenarios in<br>
> >> [1] and [2].<br>
> >><br>
> >> There is also extensive documentation on vGPU at NVIDIA's website [3].<br>
> >><br>
> >> [1] <a href="https://docs.openstack.org/nova/wallaby/admin/virtual-gpu.html" rel="noreferrer" target="_blank">https://docs.openstack.org/nova/wallaby/admin/virtual-gpu.html</a><br>
> >> [2] <a href="https://docs.openstack.org/nova/wallaby/admin/pci-passthrough.html" rel="noreferrer" target="_blank">https://docs.openstack.org/nova/wallaby/admin/pci-passthrough.html</a><br>
> >> [3] <a href="https://docs.nvidia.com/grid/13.0/index.html" rel="noreferrer" target="_blank">https://docs.nvidia.com/grid/13.0/index.html</a><br>
> >><br>
> >> Regards,<br>
> >> Gustavo.<br>
> >><br>
> >> On 17/01/2022 14:41, Satish Patel wrote:<br>
> >>> [Please note: This e-mail is from an EXTERNAL e-mail address]<br>
> >>><br>
> >>> Folk,<br>
> >>><br>
> >>> We have Tesla V100 32G GPU and I’m trying to configure with openstack wallaby. This is first time dealing with GPU so I have couple of question.<br>
> >>><br>
> >>> 1. What is the difference between passthrough vs vGPU? I did google but not very clear yet.<br>
> >>> 2. If I configure it passthrough then does it only work with single VM ? ( I meant whole GPU will get allocate to single VM correct?<br>
> >>> 3. Also some document saying Tesla v100 support vGPU but some folks saying you need license. I have no idea where to get that license. What is the deal here?<br>
> >>> 3. What are the config difference between configure this card with passthrough vs vGPU?<br>
> >>><br>
> >>><br>
> >>> Currently I configure it with passthrough based one one article and I am able to spun up with and I can see nvidia card exposed to vm. (I used iommu and vfio based driver) so if this card support vGPU then do I need iommu and vfio or some other driver to make it virtualize ?<br>
> >>><br>
> >>> Sent from my iPhone<br>
> >>><br>
> ><br>
<br>
</blockquote></div>