• Skip to primary navigation
  • Skip to main content
  • Skip to primary sidebar

Yellow Bricks

by Duncan Epping

  • Home
  • ESXTOP
  • Stickers/Shirts
  • Privacy Policy
  • About
  • Show Search
Hide Search

nvidia

Whitepaper: Running Augmented and Virtual Reality Applications on VMware vSphere using NVIDIA CloudXR

Duncan Epping · May 15, 2020 · Leave a Comment

As many of you know by now, I worked on this project with the VXR team at VMware to try to run Augmented and Virtual Reality Applications on VMware vSphere. The white paper demonstrates that, using VMware vSphere backed by NVIDIA Virtual GPU technology, AR/VR applications can be run on a Windows 10 virtual machine with an NVIDIA vGPU, and streamed to a standalone AR/VR device, such as the Oculus Quest or Vive Focus Plus, using NVIDIA’s CloudXR protocol. It was a very interesting project as we had some real challenges I did not expect. I am not going to reveal the outcome of the project and our findings, you will need to read the white paper for that, it will also give you a good understanding of the use cases around these technologies in my opinion. One thing I can reveal right here though is that these workloads are typically graphic intense. I want to share with you one image which in my opinion explains why this is:

Traditional apps/workloads usually run on a single monitor with a frame rate of 30 frames per second. VR applications are presented in a VR headset. A VR headset has a display for both eyes, that doubles the number of megapixels per second immediately, but these displays also expect 72 frames per second or more typically. This is to avoid motion sickness. All of this is described in-depth in the white paper, of course including our findings around GPU utilization when running VR/AR applications using NVIDIA CloudXR, NVIDIA and VMware vGPU on top of VMware vSphere. I hope you enjoy reading the paper as much as I enjoyed the project!

Go here to sign up for the white paper: https://pathfinder.vmware.com/activity/projectvxr

Last day of my Take 3 with the @ProjectVXR team!

Duncan Epping · Feb 28, 2020 · Leave a Comment

Today is the last day of my Take 3 with the Project VXR team. Ridiculous how fast these 3 months went. It seems like only yesterday that I posted I was going to start this journey of learning more about the world of Spatial Computing and Remote Rendering of VR in particular. If I say so myself, I feel that over the past 3 months I managed to accomplish quite a lot. The last three months I spend figuring out how to virtualize a Virtual Reality application and how to stream the application to a head-mounted display. I tested different solutions in this space, some of which I discussed on my blog in the past months, and I was surprised how smooth it worked, to be honest. If you are interested in this space, my recommendation would be to look into NVIDIA CloudXR in combination with vSphere 6.7 U3 and NVIDIA vGPU technology.

I can’t share all the details just yet, I wrote a white paper, which now needs to go through reviews and copy editing, and hopefully, it will be published soon. I will, however, discuss some of my findings and my experience during some of the upcoming VMUGs I will be presenting at. Hopefully, people will enjoy it and appreciate it.

One thing I would like to do is thank a few people who helped me tremendously in the past few months. First of all, of course, the folks on the Project VXR team, they gave me all the pointers/hints/tips I needed and shared a wealth of knowledge on the topic of spatial computing. I also like to thank Grid Factory, Ben in particular, for the many discussions, emails etc that we had. Of course also NVIDIA for the discussions and help around the lab equipment. Last but not least, I want to thank the VMware OCTO team focussing on Dell Technologies for providing me with a Dell Precision Workstation, and shipping it out literally within a day or two. Much appreciated everyone!

Now it is time to get back to reality.

NVIDIA rendering issues? Look at the stats!

Duncan Epping · Feb 14, 2020 · Leave a Comment

I’ve been down in the lab for the last week doing performance testing with virtual reality workloads and streaming these over wifi to an Oculus Quest headset. In order to render the graphics remotely, we leveraged NVIDIA GPU technology (RTX 8000 in my case here in the lab). We have been getting very impressive results, but of course at some point hit a bottleneck. We tried to figure out which part of the stack was causing the problem and started looking at the various NVIDIA stats through nvidia-smi. We figured the bottleneck would be GPU, so we looked at things like GPU utilization, FPS etc. Funny enough this wasn’t really showing a problem.

We then started looking at different angles, and there are two useful commands I would like to share. I am sure the GPU experts will be aware of these, but for those who don’t consider themselves an expert (like myself) it is good to know where to find the right stats. While whiteboarding the solution and the different part of the stacks we realized that GPU utilization wasn’t the problem, neither was the frame buffer size. But somehow we did see FPS (frames per second) drop and we did see encoder latency go up.

First I wanted to understand how many encoding sessions there were actively running. This is very easy to find out by using the following command. The screenshot below shows the output of it.

nvidia-smi encodersessions

As you can see, this shows 3 encoding sessions. One H.264 session and two H.265 sessions. Now note that we have 1 headset connected at this point, but it leads to three sessions. Why? Well, we need a VM to run the application, and the headset has two displays. Which results in three sessions. We can, however, disable the Horizon session using the encoder, that would save some resources, I tested that but the savings were minimal.

I can, of course, also look a bit closer at the encoder utilization. I used the following command for that. Note that I filter for the device I want to inspect which is the “-i <identifier>” part of the below command.

nvidia-smi dmon -i 00000000:3B:00.0

The above command provides the following output, the “enc” column is what was important to me, as that shows the utilization of the encoder. Which with the above 3 sessions was hitting 40% utilization roughly as shown below.

How did I solve the problem of the encoder bottleneck in the end? Well I didn’t, the only way around that is by having a good understanding of your workload and proper capacity planning. Do I need an NVIDIA RTX 6000 or 8000? Or is there a different card with more encoding power like the V100 that makes more sense? Figuring out the cost, performance and the trade-off here is key.

Two more weeks until the end of my Take 3 experience, and what a ride it has been. If you work for VMware and have been with the company for 5 years… Consider doing a Take 3, it is just awesome!

vGPUs and vMotion, why the long stun times?

Duncan Epping · Feb 7, 2020 · 2 Comments

Last week one of our engineers shared something which I found very interesting. I have been playing with Virtual Reality technology and NVIDIA vGPUs for 2 months now. One thing I noticed is that we (VMware) introduced support for vMotion in vSphere 6.7 and support for vMotion of multi vGPU VMs in vSphere 6.7 U3. In order to enable this, you need to set an advanced setting first. William Lam described this in his blog how to set this via Powershell or the UI. Now when you read the documentation there’s one thing that stands out, and that is the relatively high stun times for vGPU enabled VMs. Just as an example, here are a few potential stun times with various sized vGPU frame buffers:

  • 2GB – 16.5 seconds
  • 8GB – 61.3 seconds
  • 16GB – 100+ seconds (time out!)

This is all documented here for the various frame buffer sizes. Now there are a couple of things to know about this. First of all, the time mentioned was tested with 10GbE and the NVIDIA P40. This could be different for an RTX6000 or RTX8000 for instance. Secondly, they used a 10GbE NIC. If you use multi-NIC vMotion or for instance a 25GbE NIC than results may be different (times should be lower). But more importantly, the times mentioned assume the full frame buffer memory is consumed. If you have a 16GB frame buffer and only 2GB is consumed then, of course, the stun time would be lower than the above mentioned 100+ seconds.

Now, this doesn’t answer the question yet, why? Why on earth are these stun times this long? The vMotion process is described in this blog post by Niels in-depth, so I am not going to repeat it. It is also described in our Clustering Deep Dive book which you can download here for free. The key reason why with vMotion the “down time” (stun times) can be kept low is that vMotion uses a pre-copy process and tracks which memory pages are changed. In other words, when vMotion is initiated we copy memory pages to the destination host, and if a page has changed during that copy process we mark it as changed and copy it again. vMotion does this until the amount of memory that needs to be copied is extremely low and this would result in a seamless migration. Now here is the problem, it does this for VM memory. This isn’t possible for vGPUs unfortunately today.

Okay, so what does that mean? Well if you have a 16GB frame buffer and it is 100% consumed, the vMotion process will need to copy 16GB of frame buffer memory from the source to the destination host when the VM is stunned. Why when the VM is stunned? Well simply because that is the point in time where the frame buffer memory will not change! Hence the reason this could take a significant number of seconds unfortunately today. Definitely something to consider when planning on using vMotion on (multi) vGPU enabled VMs!

Disabling the frame rate limiter for your vGPU

Duncan Epping · Feb 4, 2020 · Leave a Comment

I have been testing with Virtual Reality apps within a VM for the past few days and I am leveraging NVIDIA vGPU technology on vSphere 6.7 U3. I was running into some undesired behavior and was pointed to the fact that this could be due to the frame rate being limited by default (Thank Ben!). I first checked the head-mounted display to see at what kind of frame rate it was running, by leveraging “adbLink” (for Mac) and the logcat command I could see the current frame rate hovering between 55-60. For virtual reality apps that leads to problems when moving your head from left to right as you will see black screens. For the Quest, for those wanting to play around with it as well, I used the following command to list the current frame rate for the NVIDIA CloudXR application (note that this is specific to this app) and “-s” filters for the keyword “VrApi”:

logcat -s VrApi

The result will be a full string, but the important bit is the following:

FPS=72,Prd=32ms,Tear=0,Early=0

I was digging through the NVIDIA documentation and it mentioned that if you used the Best Effort scheduler a frame rate limit would be applied. I wanted to test with the different schedulers anyway so I switched over to the Equal Share scheduler, which solved the problems listed above as it indeed disabled the frame rate limit. I could see the frame rate going up between 70 and 72. Of course, I also wanted to validate the “best effort” scheduler with frame rate limit disabled, I did this by adding an advanced setting to the VM:

pciPassthru0.cfg.frame_rate_limiter=0

This also resulted in a better experience, and again the frame rate going up to 70-72.

  • Go to page 1
  • Go to page 2
  • Go to Next Page »

Primary Sidebar

About the author

Duncan Epping is a Chief Technologist in the Office of CTO of the Cloud Platform BU at VMware. He is a VCDX (# 007) and the author of the "vSAN Deep Dive" and the “vSphere Clustering Technical Deep Dive” series.

Upcoming Events

04-Mar-21 | Polish VMUG – Roadshow
09-Mar-21 | Austria VMUG – Roadshow
16-Mar-21 | VMUG Turkey – Roadshow
18-Mar-21 | St Louis Usercon Keynote
26-Mar-21 | Hungary VMUG – Roadshow
08-Apr-21 | VMUG France – Roadshow

Recommended reads

Sponsors

Want to support us? Buy an advert!

Advertisements

Copyright Yellow-Bricks.com © 2021 · Log in