NVIDIA GRID 4.2 released (367.92/369.95)

Written by Thomas Poppelgaard. Posted in CentOS, Citrix, Datacenter and Cloud, Desktop Virtualization, End User Computing, GRID, HDX, HDX 3D, HDX 3D Pro, Horizon, Linux, Microsoft, NVENC, NVIDIA, Quadro, RHEL, Tesla, Ubuntu, vDGA, vGPU, VMware, vSphere, Windows 10, Windows 7, Windows 8, Windows 8.1, Windows Server 2008R2, Windows Server 2012R2, Windows Server 2016, XenApp, XenDesktop, XenServer

nvidiagrid

Hi All

Its time to update your NVIDIA GRID K1, K2, M10, M6, M60 environment.

NVIDIA have released new drivers for NVIDIA GRID 4.2 for March 2017.

New in this Release:

  • NVIDIA GRID 4.2 has multiple fixes that improve memory management and NVFBC optimizations that ensure that surface allocations in the Frame Buffer are managed more efficiently to increase stability. The latest release includes frame buffer residency optimizations. Now, during high-memory pressure situations where there is a lack of or high competition for frame buffer, users remain resident-in.
  • Miscellaneous bug fixes
  • Support for Citrix XenServer 7.1
  • Support for the following Linux guest OS versions:
    • Red Hat Enterprise Linux 7.3
    • CentOS 7.3
    • Ubuntu 16.04 LTS

Important notes – Note: XenServer 7.1 is the last XenServer release supported on GRID K1 and K2. Citrix XenServer 6.2 is no longer supported.

Whats new in vGPU 367.92-369.95

NVIDIA have released a new version of vGPU 367.92-369.95 for NVIDIA GRID  (K1, K2, Tesla M6, M10 and M60 platform)

Included in this release is

  • NVIDIA GRID Virtual GPU Manager versions 367.92 for Citrix XenServer 6.5 SP1
  • NVIDIA GRID Virtual GPU Manager versions 367.92 for Citrix XenServer 7
  • NVIDIA GRID Virtual GPU Manager versions 367.92 for Citrix XenServer 7.1
  • NVIDIA GRID Virtual GPU Manager version 367.92 for VMware vSphere 6.0 Hypervisor (ESXi)
  • NVIDIA GRID Virtual GPU Manager version 367.92 for VMware vSphere 6.5 Hypervisor (ESXi)
  • NVIDIA Windows drivers for vGPU version 369.95
  • NVIDIA Linux drivers for vGPU version 367.92.

Important:

The GRID vGPU Manager and Windows guest VM drivers must be installed together. Older VM drivers will not function correctly with this release of GRID vGPU Manager. Similarly, older GRID vGPU Managers will not function correctly with this release of Windows guest drivers

Windows Guest OS support in vGPU 369.95

GRID vGPU 369.95 supports following Windows release as a guest OS

  • Microsoft Windowss 7 (32/64bit)
  • Microsoft Windows 8 (32/64bit)
  • Microsoft Windows 8.1 (32/64bit)
  • Microsoft Windows 10 (32/64bit)
  • Microsoft Windows Server 2008R2
  • Microsoft Windows Server 2012 R2
  • Microsoft Windows Server 2016

Linux Guest OS support in vGPU 367.92

GRID vGPU 367.92 supports following Linux distributions as a guest OS only on supported Tesla GPUs
Import: GRID K1/K2 do not support vGPU on Linux guest os.

  • Red Hat Enterprise Linux 7.0-7.3
  • CentOS 7.0-7.3
  • Ubuntu 16.04 LTS
  • Ubuntu 14.04 LTS

Guide – Update existing vGPU GRID Manager (Hypervisor)

 

GRID vGPU Manager 367.92 for Citrix XenServer 6.5

If you have a NVIDIA GRID K1, K2, M6, M60 vGPU GRID manager installed in Citrix XenServer. Upgrade with one of below methodology:

Methodology 1 – the manual way “No GUI”

Upgrading an existing installation of the NVIDIA driver on Citrix XenServer 6.5, use the rpm -U command to upgrade:

If you have NVIDIA GRID K1 / K2
[root@localhost ~]# rpm -Uv NVIDIA-vGPU-kepler-xenserver-6.5-367.92.x86_64.rpm 
Preparing packages for installation...
If you have NVIDIA GRID TESLA M10 / M6 / M60
[root@localhost ~]# rpm -Uv NVIDIA-vGPU-xenserver-6.5-367.92.x86_64.rpm 
Preparing packages for installation...

The recommendation from NVIDIA is to shutdown all VMs using a GPU. The machine does continue to work during the update, but since you need to reboot the XenServer itself, it’s better to gracefully shutdown the VMs. So after your VMs have been shutdown and you upgraded the NVIDIA driver, you can reboot your host.

[root@localhost ~]# xe host-disable
[root@localhost ~]# xe host-reboot

Methodology 2 – the “GUI” way

Select Install Update… from the Tools menu
 Click Next after going through the instructions on the Before You Start section
 Click Add on the Select Update section and open NVIDIA’s XenServer Supplemental Pack ISO

If you have NVIDIA GRID K1 / K2 select following file:

“NVIDIA-vGPU-kepler-xenserver-6.5-367.92.x86_64.iso ”

If you have NVIDIA GRID M10/ M6/M60 select following file:

“NVIDIA-vGPU-xenserver-6.5-352.83.x86_64.iso ”

Click Next on the Select Update section
 In the Select Servers section select all the XenServer hosts on which the Supplemental Pack should be installed on and click Next
 Click Next on the Upload section once the Supplemental Pack has been uploaded to all the XenServer hosts
Getting Started
 Click Next on the Prechecks section
 Click Install Update on the Update Mode section
 Click Finish on the Install Update section

After the XenServer platform has rebooted, verify that the GRID package installed and loaded correctly by checking for the NVIDIA kernel driver in the list of kernel loaded modules.

Validate from putty or XenCenter CLI

run lsmod | grep nvidia

Verify that the NVIDIA kernel driver can successfully communicate with the GRID physical GPUs in your system by running the nvidia-smi command, which should produce a listing of the GPUs in your platform:

Check driver version is 367.92, if it is then your host is ready for GPU awesomeness and make your VM rock.

GRID vGPU Manager 367.92 for Citrix XenServer 7

If you have a NVIDIA GRID K1, K2, M6, M60 vGPU GRID manager installed in Citrix XenServer. Upgrade with one of below methodology:

Methodology 1 – the manual way “No GUI”

Upgrading an existing installation of the NVIDIA driver on Citrix XenServer 7, use the rpm -U command to upgrade:

If you have NVIDIA GRID K1 / K2
[root@localhost ~]# rpm -Uv NVIDIA-vGPU-kepler-xenserver-7.0-367.92.x86_64.rpm 
Preparing packages for installation...
If you have NVIDIA GRID TESLA M10 / M6 / M60
[root@localhost ~]# rpm -Uv NVIDIA-vGPU-xenserver-7.0-367.92.x86_64.rpm 
Preparing packages for installation...

The recommendation from NVIDIA is to shutdown all VMs using a GPU. The machine does continue to work during the update, but since you need to reboot the XenServer itself, it’s better to gracefully shutdown the VMs. So after your VMs have been shutdown and you upgraded the NVIDIA driver, you can reboot your host.

[root@localhost ~]# xe host-disable
[root@localhost ~]# xe host-reboot

Methodology 2 – the “GUI” way

Select Install Update… from the Tools menu
 Click Next after going through the instructions on the Before You Start section
 Click Add on the Select Update section and open NVIDIA’s XenServer Supplemental Pack ISO

If you have NVIDIA GRID K1 / K2 select following file:

“NVIDIA-vGPU-kepler-xenserver-7.0-367.92.x86_64.iso ”

If you have NVIDIA GRID M10/ M6/M60 select following file:

“NVIDIA-vGPU-xenserver-7.0-352.83.x86_64.iso ”

Click Next on the Select Update section
 In the Select Servers section select all the XenServer hosts on which the Supplemental Pack should be installed on and click Next
 Click Next on the Upload section once the Supplemental Pack has been uploaded to all the XenServer hosts
Getting Started
 Click Next on the Prechecks section
 Click Install Update on the Update Mode section
 Click Finish on the Install Update section

After the XenServer platform has rebooted, verify that the GRID package installed and loaded correctly by checking for the NVIDIA kernel driver in the list of kernel loaded modules.

Validate from putty or XenCenter CLI

run lsmod | grep nvidia

Verify that the NVIDIA kernel driver can successfully communicate with the GRID physical GPUs in your system by running the nvidia-smi command, which should produce a listing of the GPUs in your platform:

Check driver version is 367.92, if it is then your host is ready for GPU awesomeness and make your VM rock.

GRID vGPU Manager 367.92 for Citrix XenServer 7.1

If you have a NVIDIA GRID K1, K2, M6, M60 vGPU GRID manager installed in Citrix XenServer. Upgrade with one of below methodology:

Methodology 1 – the manual way “No GUI”

Upgrading an existing installation of the NVIDIA driver on Citrix XenServer 7.1, use the rpm -U command to upgrade:

If you have NVIDIA GRID K1 / K2
[root@localhost ~]# rpm -Uv NVIDIA-vGPU-kepler-xenserver-7.1-367.92.x86_64.rpm 
Preparing packages for installation...
If you have NVIDIA GRID TESLA M10 / M6 / M60
[root@localhost ~]# rpm -Uv NVIDIA-vGPU-xenserver-7.1-367.92.x86_64.rpm 
Preparing packages for installation...

The recommendation from NVIDIA is to shutdown all VMs using a GPU. The machine does continue to work during the update, but since you need to reboot the XenServer itself, it’s better to gracefully shutdown the VMs. So after your VMs have been shutdown and you upgraded the NVIDIA driver, you can reboot your host.

[root@localhost ~]# xe host-disable
[root@localhost ~]# xe host-reboot

Methodology 2 – the “GUI” way

Select Install Update… from the Tools menu
 Click Next after going through the instructions on the Before You Start section
 Click Add on the Select Update section and open NVIDIA’s XenServer Supplemental Pack ISO

If you have NVIDIA GRID K1 / K2 select following file:

“NVIDIA-vGPU-kepler-xenserver-7.1-367.92.x86_64.iso ”

If you have NVIDIA GRID M10/ M6/M60 select following file:

“NVIDIA-vGPU-xenserver-7.1-352.83.x86_64.iso ”

Click Next on the Select Update section
 In the Select Servers section select all the XenServer hosts on which the Supplemental Pack should be installed on and click Next
 Click Next on the Upload section once the Supplemental Pack has been uploaded to all the XenServer hosts
Getting Started
 Click Next on the Prechecks section
 Click Install Update on the Update Mode section
 Click Finish on the Install Update section

After the XenServer platform has rebooted, verify that the GRID package installed and loaded correctly by checking for the NVIDIA kernel driver in the list of kernel loaded modules.

Validate from putty or XenCenter CLI

run lsmod | grep nvidia

Verify that the NVIDIA kernel driver can successfully communicate with the GRID physical GPUs in your system by running the nvidia-smi command, which should produce a listing of the GPUs in your platform:

Check driver version is 367.92, if it is then your host is ready for GPU awesomeness and make your VM rock.

GRID vGPU Manager 367.92 for VMware vSphere 6.0

To update the NVIDIA GPU VIB, you must uninstall the currently installed VIB and install the new VIB.

To uninstall the currently installed VIB:

  1. Stop all virtual machines using 3D acceleration.
  2. Place the ESXi host into Maintenance mode.
  3. Open a command prompt on the ESXi host.
  4. Stop the xorg service by running the command:/etc/init.d/xorg stop
  5. Remove the NVIDIA VMkernel driver by running the command:vmkload_mod -u nvidia
  6. Identify the NVIDIA VIB name by running this command:esxcli software vib list | grep NVIDIA
  7. Remove the VIB by running the command:esxcli software vib remove -n nameofNVIDIAVIBYou can now install a new NVIDIA GPU VIB
  8. Use the esxcli command to install the vGPU Manager package:
If you have NVIDIA GRID K1 / K2 select following file:
[root@lesxi ~] esxcli software vib install -v /NVIDIA-vGPU-kepler-VMware_ESXi_6.0_Host_Driver_367.92-1OEM.600.0.0.2494585.vib
If you have NVIDIA GRID TESLA M10 / M6 / M60 select following file:
[root@lesxi ~] esxcli software vib install -v /NVIDIA-vGPU-VMware_ESXi_6.0_Host_Driver_367.92-1OEM.600.0.0.2494585.vib

After the ESXi host has rebooted, verify that the GRID package installed and loaded correctly by checking for the NVIDIA kernel driver in the list of kernel loaded modules.

[root@lesxi ~]# vmkload_mod -l | grep nvidia 
Preparing packages for installation...

Validate

run nvidia-smi

Verify that the NVIDIA kernel driver can successfully communicate with the GRID physical GPUs in your system by running the nvidia-smi command, which should produce a listing of the GPUs in your platform:

Check driver version is 367.92, if it is then your host is ready for GPU awesomeness and make your VM rock.

GRID vGPU Manager 367.92 for VMware vSphere 6.5

To update the NVIDIA GPU VIB, you must uninstall the currently installed VIB and install the new VIB.

To uninstall the currently installed VIB:

  1. Stop all virtual machines using 3D acceleration.
  2. Place the ESXi host into Maintenance mode.
  3. Open a command prompt on the ESXi host.
  4. Stop the xorg service by running the command:/etc/init.d/xorg stop
  5. Remove the NVIDIA VMkernel driver by running the command:vmkload_mod -u nvidia
  6. Identify the NVIDIA VIB name by running this command:esxcli software vib list | grep NVIDIA
  7. Remove the VIB by running the command:esxcli software vib remove -n nameofNVIDIAVIBYou can now install a new NVIDIA GPU VIB
  8. Use the esxcli command to install the vGPU Manager package:
If you have NVIDIA GRID K1 / K2 select following file:
[root@lesxi ~] esxcli software vib install -v /NVIDIA-vGPU-kepler-VMware_ESXi_6.5_Host_Driver_367.92-1OEM.650.0.0.2494585.vib
If you have NVIDIA GRID TESLA M10 / M6 / M60 select following file:
[root@lesxi ~] esxcli software vib install -v /NVIDIA-vGPU-VMware_ESXi_6.5_Host_Driver_367.92-1OEM.650.0.0.2494585.vib

After the ESXi host has rebooted, verify that the GRID package installed and loaded correctly by checking for the NVIDIA kernel driver in the list of kernel loaded modules.

[root@lesxi ~]# vmkload_mod -l | grep nvidia 
Preparing packages for installation...

Validate

run nvidia-smi

Verify that the NVIDIA kernel driver can successfully communicate with the GRID physical GPUs in your system by running the nvidia-smi command, which should produce a listing of the GPUs in your platform:

Check driver version is 367.92, if it is then your host is ready for GPU awesomeness and make your VM rock.

Update existing vGPU Driver (Virtual Machine)

When the hypervisor vGPU GRID manager is updated next is updating the Virtual Machines vGPU driver.

  • 369.95_grid_win8_win7_32bit_international.exe
  • 369.95_grid_win8_win7_server2012R2_server2008R2_64bit_international.exe
  • 369.95_grid_win10_32bit_international.exe
  • 369.95_grid_win10_server2016_64bit_international.exe
  • NVIDIA-Linux-x86_64-367.92-grid.run (only available with M10/M6/M60)

The vGPU driver for Windows 7, 8, 8.1, 10 is available with NVIDIA GRID vGPU download. This is available for both K1/K2/M10/M6/M60

The vGPU driver for Linux is only available for vGPU with following Tesla GPU M10/M6/M60

Update your Golden Images and reprovisioning the new virtual machines with updated vGPU drivers, if you have stateless machines update vGPU drivers on each.

#HINT – Express upgrade of drivers is the recommended option according to the setup. If you use the “Custom” option, you will have the option to do a “clean” installation. The downside of the “clean installation” is that it will remove all profiles and custom settings. The pro of using the clean installation option is that it will reinstall the complete driver, meaning that there will be no old driver files left on the system. I most of the time recommends using a “Clean” installation to keep it vanilla 🙂

#HINT (Citrix XenDesktop 7.6-7.11 customers)

The NVIDIA GRID API provides direct access to the frame buffer of the GPU, providing the fastest possible frame rate for a smooth and interactive user experience. If you install NVIDIA drivers before you install a VDA with HDX 3D Pro, NVIDIA GRID is enabled by default.

To enable NVIDIA GRID on a VM, disable Microsoft Basic Display Adapter from the Device Manager. Run the following command and then restart the VDA: Montereyenable.exe -enable -noreset

If you install NVIDIA drivers after you install a VDA with HDX 3D Pro, NVIDIA GRID is disabled. Enable NVIDIA GRID by using the Montereyenable tool provided by NVIDIA.

To disable NVIDIA GRID, run the following command and then restart the VDA: Montereyenable.exe -disable -noreset

#HINT (Citrix XenDesktop 7.12/7.13 customers)

The NVIDIA GRID API provides direct access to the frame buffer of the GPU, providing the fastest possible frame rate for a smooth and interactive user experience. If you install NVIDIA drivers before you install a VDA with HDX 3D Pro, NVIDIA GRID is enabled by default.

To enable NVIDIA GRID on a VM, disable Microsoft Basic Display Adapter from the Device Manager. Run the following command and then restart the VDA: NVFBCEnable.exe -enable -noreset

If you install NVIDIA drivers after you install a VDA with HDX 3D Pro, NVIDIA GRID is disabled. Enable NVIDIA GRID by using the NVFBCEnable tool provided by NVIDIA.

To disable NVIDIA GRID, run the following command and then restart the VDA: NVFBCEnable.exe -disable -noreset

 

Source

GRID K1/K2 – sources

Download vGPU 352.83-354.80 for XenServer 6.5 for GRID K1 / K2 here

Download vGPU 352.83-354.80 for XenServer 7.0 for GRID K1 / K2 here

Download vGPU 367.92-369.95 for XenServer 7.1 for GRID K1 / K2 here

Download vGPU 367.92-369.95 for vSphere 6 for GRID K1 / K2 here

Download vGPU 367.92-369.95 for vSphere 6.5 for GRID K1 / K2 here

Tesla M10/M6/M60 – sources

vGPU Grid Manager + Drivers are only available to customers and NVIDIA NPN partners for M10/M6/M60.

Download if you are a NPN partner

Download if you are a GRID M10, M6, M60 customer

vGPU 367.92-369.95 for XenServer 6.5 for Tesla M10 / M6 / M60

vGPU 367.92-369.95 for XenServer 7 for Tesla M10 / M6 / M60

vGPU 367.92-369.95 for XenServer 7.1 for Tesla M10 / M6 / M60

vGPU 367.92-396.95 for vSphere 6 for Tesla M10 / M6 / M60

vGPU 367.92-396.95 for vSphere 6.5 for Tesla M10 / M6 / M60

Other Sources

NVIDIA blog – https://blogs.nvidia.com/blog/2017/03/10/nvidia-grid-march-2017/ 

HPE Discover London 2015

Written by Thomas Poppelgaard. Posted in Augmented Reality, Citrix, Cloud, Desktop Virtualization, End User Computing, Envy, Events, FrameHawk, GRID, GRID 2.0, HDX, HDX 3D, HDX 3D Pro, HPE, HPE Discover, M710P, Moonshot, NVIDIA, Oculus Rift, Provisioning Services, Quadro, Sprout, Tesla, Virtual Reality, VMware, XenDesktop, XenServer

HPdiscover-independentInflucerprogram

Last year I was invited by HPE to go to their EMEA event HPE Discover in London, which took place the first week of December. I was invited there as part of their Hewlett Packard Enterprise Independent Influencer Program, and I was super proud of being part of this well organized program, which Kristen Reyes manages. HUGE kudos to Kristen Reyes and her team for putting together one of the most organized events I have ever attended as a blogger. Last time I attended HPE Discover was some years ago at their big US HPE Discover which took place at Las Vegas. I remember it was HUGE with lots of people – an epic experience that my previously employer, Arrow ECS managed the trip for. I had pretty high expectations going to the HPE Discover and I knew it was going to be good, but not as good as it REALLY was.

HPE Independent Influencer Program:

I am proud of being part of this program – let me tell you which profiles are part of the program including myself.

The team behind HPE took care of us from the time I flew all the way from Denmark to London. They made sure we got the right hotel and then once at the venue, we got a special VIP room where we got introduce each other. Team had a plan for what happend exclusively for the group. This reminds me of the Citrix CTP program I am humble to be part of where we also meet 2 times a year and are taken especially care of.

HPdiscoverLondon2015-3

If you want to read more about the individuals tech bloggers, that are in the program, click here.

I highly recommend you follow above persons, as they are awesome people. This is the first time I have been with such a wide  range of technology bloggers in my life and it was a huge difference – so thank you to HPE for daring to put so many different profiles together.

 

Let’s talk about HPE Discover in London 2015

There was 13.000 attendees. YES! That is INSANE and it felt like being in United States but I was in London. London can also have huge venues and they succeeded with cherry top on the beautiful cake. Yes HPE pulled it off! Please other big vendors – start making EMEA events because London is a perfect spot for this! Citrix, VMware, Microsoft… EMEA salutes you..

poppelgaard_twitter01

HPdiscoverLondon2015

I was in for a surprise the first day of the keynote when Meg Whitman kicked off HPE Discover, and Satya Nadella Microsoft CEO talked even thought it was via a conf call (but still pretty cool) and Microsoft showed the flag. I would have loved if Satya could have come in person since that would have been my first time seeing Mr. Nadella in person.

poppelgaard_twitter02

After a great keynotes learning about where HPE is taking the company, it was time to walk around the expo to see what innovative technologies were hidden and what others I was allowed to blog about.

Next gen consumer products for engineers & designers

HP ENVY 34

HP envy 4k IMG_5222

This device is so beautiful the brand new HP Envy 34, its the widest screen I have ever seen on a professional computer and its powered by Intel HD530 GPU and NVIDIA GTX 960A GPU. The speakers are B&O (Danish design brand).

Price for this device is approx 1700 USD – that’s pretty cheap for an All-in-one machine.

Another beautiful Augmented Reality device I tried was the HP Sprout. This device makes it possible to quickly edit 3D models in real time by taking objects from reality and scanning them directly in to the device. Then you can manipulate with the object and then print it out on a 3Dprinter. This process increases  the speed of making a design. Right now, the device is only available for consumers but we might see that change soon for enterprises since the unit was displayed at the HPE Discover 🙂  – so watch out for this space. I also saw AMD had setup several Oculus Rift DK2 units running some great demos. Great to see that VR was at HPE Discover.

HP sprout HPdiscoverLondon2015-VR

Deep Learning & The Machine

IMG_5231  the machine

I meet many of the brains behind the Deep Learning “things” HPE are doing, and its VERY impressive where they are, and how HPE differs to their competitors. I can’t talk much about it now – and will leave it up to how things are evolving later this year – and then blog more about it when the information is available to the public. But if you are fascinated by the future and AI / Deep learning you should look at what HPE is doing with the concept, “The Machine”. If you look at what they are trying to accomplish and put the puzzle together yourself, you might see for yourself how HPE will make a difference in how next generation datacenters will be built – that’s a fact.

Learn more about the HPE “The Machine” here

Citrix Octoblu & HPE – The Cube
(The ultimate iOT automated conference room)

The Cube

HPE Smart Digital Collaboration Space (1107), a.k.a. “The Cube.” . That was the name of this cool concept.

This was mind-blowing – and one of the most innovative things I saw at HPE Discover with IoT using Citrix Octoblu.

HPE, together with Aruba and Octoblu have teamed together to create a beautiful workflow that automates a conference room with iOT.

All visitors could try this at the event. HPE allowed 3-4 people to enter the cube as a group. We were asked our name and then were given an iPhone to bring into “The Cube”. The system then welcomed use by name and the room went from “green” to “red” (which was Philips Hue that did this), so the people outside could clearly see that the room was occupied and a meeting was taking place. An HP tablet on the door outside also showed the room was occupied and listed the attendees by name. Inside, the HP surface PCs connected to Skype for Business, Samsung TV started and a business session went live. At the end,  “The Cube” took a selfie and put on twitter and sent the photo to our emails and after we left the room. Upon exiting, it reset all the workflows and the room was lit green again – ready for next group. Amazing 🙂

IMG_5255IMG_5258

Below is the workflow that was created in Octoblu

Octoblue Workflow HPE

Try out octoblu here if you havent tried it yet. Its still beta and “FREE” https://octoblu.com/

Read the official blog post from Citrix about the solution

 

HPE & NVIDIA GRID offerings

HPDiscoverLondon2015-NVIDIAnvidia-tesla-m6

It was great to see all the new NVIDIA GRID 2.0 solutions. Two gentlemen proudly showed the HPE ProLiant WS460C Gen9 that now supports NVIDIA Tesla M60 and NVIDIA Tesla M6. This means higher density and more performance. So great to see that HPE is a pioneer, offering the entire solution from NVIDIA. There is soon NVIDIA GTC – and I expect we will see some data on customers running on HPE solution with M60/M6 and we get some data on how many users can run on these new offerings. It’s actually my first time seeing on-hands the M6 GPU from NVIDIA and it looks beautiful and green. These GPUs are made for low end GPU workloads all the way up to high end GPU workloads. The performance of the M60 is 2x the performance compared to K2 (GRID 1.0)

Please keep in mind that NVIDIA has changed GRID 2.0 so when you buy the GPUs, you also have to buy a “software” license.
Your NPN (NVIDIA preferred partner) should be able to help you with providing you the correct information. This license is required both for Pass-through and vGPU of the M60 and M6.

grid-license

 

Moonshot and beyond

IMG_5275IMG_5276IMG_5278

HPE Moonshot has been a subject I have been following for a while now. It all started when one of my great friends (and fellow CTP) Dane Young talked about HP Moonshot and why it differs to the market. HPE liked Young’s view – and his face is now on the wall at the Houston research facility. So if you are a client or a partner, you might know Dane Young – or you will if you are in Houston to see what HPE is doing with HPE Moonshot.

Citrix had a booth at the HPE Discover and they showed how Citrix is working and offered on a HPE Moonshot stack.

The HPE Moonshot team had several sessions talking about the next generation “trader workstation” which opens my eyes on how users use a computer. Traders are actually some of the most “difficult” users to satisfy – and they have a very short lifetime as an employee because its so hard what they do is so stressful. The amount of monitors and amount of data they look at is shocking and the high availability for this is nothing like you have never seen before. If these people cant work, the moneyflow stops moving – something you don’t want in the financial world. Designers are one thing, but I would definitely call traders the most difficult users. So what´s so amazing about the HPE Moonshot offering? HPE has a new catridge called the M710P, which was released in the Summer 2015. It has a high amount of CPU/GPU (or as others call it APU which is a CPU/GPU processor in 1 unit). Lots of memory and high speed disks. This is all build in a super small server which we call a cartridge. This cartridge can be put in a system and an OS can be installed and now you can have many users connecting to the catridge and using multiple sessions from Microsoft Remote Desktop Services combine this with Citrix super power “XenApp” and you now have an amazing flagship for having high amount of users with low cost. Nothing is virtualized and everything is running on bare metal. The beauty about the moonshot is that in 1 chassis you can put in 45 catridges of the M710P.

Moonshot cartridge building blocks for Citrix workloads

These are the offering that HPE have with Moonshot and Citrix and you can build amazing workloads. Personally, I would look at the M710p because I think this is the cartridge that is mature for the workload. I normally design Citrix solutions with.. wait are you talking about GPU? Yes I am..

moonshot-buildingblocks

If you remember some time ago, Brian Madden, said “Imagine if you could put 20.000 users in a single rack? Let’s see what has happended since then. Things have evolved – and with the new cartridge M710p, its much more powerful than the first cartridges that were built for Citrix

brianmadden_moonshot

Citrix is using Moonshot internally

At HPE Discover, Citrix had a session with Bob Thompson from Citrix and together with the HPE Moonshot team EMEA telling how Citrix is using HPE Moonshot.

I had never heard about Citrix is using it at this scale – and big surprise their demo cloud is also running on Moonshot.

poppelgaard_twitter03

Insights what Citrix IT are doing with HPE Moonshot

  • “Citrix on Citrix” – team responsible for deploying Citrix products internally
  • Upgrading to latest version of XenApp
  • Seamless app access from Windows 10 devices
  • Superior user experience
  • Improved security (e.g. both FIPS Compliant and Common Criteria certified)
  • Migrating to Moonshot (m710) for XenApp from HP Blades
  • Designed and optimized for XenApp
  • Less: power consumption, space and cabling
  • 8,500 sessions, across the globe in (4) data centers

“WOW so why arent more people using Moonshot, if this is so good and Citrix is doing this, it is mature”.

citrix-moonshot

citrix-moonshot-02

M710P & Citrix – Next gen trader workstation

I am very impressed about this new machine and HPE is having HUGE success with this machine for the financial buisness. In fact, some of the biggest banks in the world are running on this. Surprised ? IT’S TRUE

Let’s look at the details of what it is.

moonshot-m710p-nextgentraderworkstation

Here is a typical trader running with 4 monitors connecting to 1 catridge M710p and only 47% of the GPU is utilized. AMAZING!

IMG_5246 IMG_5245

HPE-nextgentraderworkstation

 

M710P catridge – hardware specifications

m710p front m710p back

WORKLOADs Video transcoding , Application Delivery
CPU+GPU Intel Xeon E3-1284Lv4 with on-die GPU (GT3e: Iris Pro P6300)

4 core, 8 thread, 2.9 GHz (3.8GHz Turbo), GPU, 128MB DRAM

MEMORY 32GiB of ECC-protected memory, dual-memory channels

4 x 8GB LV SO-DIMMs at 1600MT/s

NETWORK Integrated NIC: dual port 10GbE Mellanox CX3 PRO

Supported Switch(s): 45 port 10GbE Downlinks, 4 x 40GbE QSFP uplinks

STORAGE Local SSD boot and data: 120GB, 240GB, 480GB or 960GB

m.2 (2280) form factor

POWER Cartridge: 83W peak, 50W typical
OS Ubuntu 15.05 w/KVM, RHEL 6.5,7.0 w/KVM, SLES 11.4/12,

Windows Server 2012 R2, 2008R2, CentOS 6.7, 7.2

What you get in a 4.5U chassis?

  • 1.44TB RAM
  • 90 10Gb NICs (internal)
  • 8 x 40Gb QSFP+ uplinks (320Gb/s total bandwidth)
  • 2 independent L2/3 switches with 450Gb throughput
  • 45 GPUs
  • 45 CPUs
  • 180 Cores (360 with HyperThreading)
  • 5.4TB, 10.8Tb, 21.6TB or 43TB iSSD Storage
  • 1800 users (conservative @ 40 per server)

Moonshot – what can you get in a rack

  • 10 x 4.3U Chassis
  • 14.4TB RAM
  • 20 independent L2/3 switches with 4.5Tb throughput
  • 450 GPUs
  • 450 CPUs
  • 1180 Cores (3600 with HyperThreading)
  • 54TB – 430TB iSSD Storage
  • 18,000 XenApp users (@40 users per server)

I hope the above gives you a pretty good idea on where Moonshot is going – and this is just for Citrix. HPE Moonshot is much more though -it’s a huge eco system build for enterprises and cloud providers, so check it out https://www.hpe.com/us/en/servers/moonshot.html

More informations about HPE Moonshot M710p – The Trader Workstation:

Bringing the Benefits of a Hosted Desktop Solution to Traders

Where can I try this

Book at demo and try HP Moonshot M700/M710/M710p here –  http://discoverylab-hpe.com/book-a-demo/

 

Summary

HPE Discover 2015 blew my mind – and it looks like we are going to have an amazing year in 2016.

I would like to say thank you to Kristen Reyes from HPE for making this happen. Each time I didn’t know what to do, she and her team took great care of me and made sure I meet the correct people at HPE – and there was always “open doors” (which isn’t always the case when techies sees a guy with a “blogger” sign around his neck).

Kristen Reyes

 

DEVELOP3D LIVE, Warwick, 31March 2016

Written by Thomas Poppelgaard. Posted in AMD, Augmented Reality, Autodesk, Azure, Citrix, Datacenter and Cloud, Desktop Virtualization, DEVELOP3DLive, DirectX 12, End User Computing, Events, FrameHawk, GRID, GRID 2.0, HDX 3D, HDX 3D Pro, HP, Hyper-V, Intel, Leap Motion, Linux, Microsoft, NVIDIA, Oculus Rift, Quadro, RemoteFX, vGPU, Virtual Reality, VMware, Windows 10, Windows 7, Windows 8.1, Windows Server 2012R2, Windows Server 2016, XenApp, XenDesktop

develop3dlive

Hi all, if you are looking for an event that is all about CAD, CAM, CAE, PLM, Rendering, 3D technologies, Augmented Reality, Virtual Reality and 3D printing then is FREE!! then DEVELOP3D LIVE is the event to  join. Rachel Berry opened my eyes going there last year 2015 and she made me come all the way from Denmark to Warwick and it was worth the long trip 🙂 If you are looking for travel advice Hotels and such I recommend you read Rachel Berry blogpost, I use myself Kayak.com 🙂 that always saves my life when i am traveling around the world and if you want it cheap use Airbnb.com.

My things learned from DEVELOP3D LIVE 2015

I went there and had high expectations and I was in for a surprise. It was all about above and much more  I saw for the first time Carl Bass the CEO of Autodesk, Gian Paolo Bassi the CEO of SolidWorks, Jon Hirschtick the CEO of Onshore and the founder of SolidWorks, Nikola Bozinovic the Founder and CEO of FRAME, and many many more.

I got my hands on the HP Zvr for the second time, I saw the unit for the first time at NVIDIA GTC in 2013 and HP have acquired the company. I also saw another favorite Augmented Device the HP Sprout, which is an amazing unit. 3Dconnexion was off course also there so all my favorite CAD peripherals and next gen tools.

Below is the picture of Carl Bass CEO of Autodesk talking at DEVELOP3D LIVE last year and Rachel Berry trying Fra.me solution

Carl Bass CEO AutodeskRachel Berry trying out Fra.me

Meet the Speakers of DEVELOP3D LIVE 2016

I am looking very much forward meeting a good friend of mine Adam Jull from IMSCAD Global.
Adam and his team have been early adapters of HDX 3D Pro like myself, so its great to see a pioneer talking at Develop3DLive.

d3dlivespeakers1

d3dlivespeakers

I am also looking very much forward meeting the NVIDIA team + my favorite CAD genius Rachel Berry and also see if AMD have something interesting to show like the Multiuser gpu if its ready for that time.

 

Get your ticket for free here for joining the awesome CAD event in beautiful Warwick. If you never been to UK well this is your reason to go, learn from the best and brightest people on CAD and see the beautiful country side of England. Two awesome things.

AutoCAD mouse pointer issues with Citrix XenApp 7.6

Written by Thomas Poppelgaard. Posted in AMD, AutoCAD, Autodesk, Citrix, FirePro, GRID, GRID 2.0, HDX 3D, HDX 3D, Intel, Microsoft, NVIDIA, Quadro, vGPU, Windows Server 2008R2, Windows Server 2012, Windows Server 2012R2, XenApp

Citrix have released a limited hotfix for Citrix XenDesktop 7.6 and XenApp 7.6 that fixes an issue with Autodesk AutoCAD & AutoCAD Civil 3D where the  mouse pointer permanently displays as an hour glass. By applying this hotfix this fixes the issue for AutoCAD.

Source

Download the ICATS760WX64039 – For VDA Core Services 7.6 for Windows Server OS (64-bit) for Citrix XenApp 7.6 here

 

 

Webinar I did with XenAppblog – “GPU in virtualization, learn why it’s important” now available

Written by Thomas Poppelgaard. Posted in 3DConnexion, AMD, Apple, Autodesk, Best Practise, Cisco, Citrix, Dell, FirePro, GRID, HDX 3D, HDX 3D Pro, HowTo, HP, HTML5, IBM, Lakeside Software, LoginVSI, Microsoft, NVIDIA, Quadro, Receiver, Reciever, RemoteFX, Server 2012R2, Splunk, UberAgent, vDGA, vGPU, View, Virtualization Ready, VMware, vSGA, vSphere, Whitepapers, Windows 7, Windows 8, Windows 8.1, Windows Server 2008R2, Windows Server 2012, Wyse, XenApp, XenDesktop, XenServer

Hi All

I am very excited to share this great news with you all. I did a webinar with fellow CTP  Trond Eirik Håvarstein from XenAppBlog.com, and we had a special guest surprise Jeroen Van De Kamp CTP and CTO, LoginVSI announcing ground breaking stuff in the webinar. We had over 700 people signed up for the Webinar, if you was among the crowd that missed the opportunity to see the webinar here is your chance, the webinar is now available for everyone for free. There was a lot of Q/A and I will the next couple of days reply to all the Q/A and make them available in this article.

The webinar has been re-mastered and the audio & graphical demo videos is even better now  than in the actual webinar, make sure to check it out now:

Download the presentation here (PDF format)

Summary of webinar product announcements from LoginVSI, Lakeside Software, Uberagent for Splunk.

loginvsi
LoginVSI upcoming new version support’s GPU benchmark…

LoginVSI is working on next version that will support benchmark, capacity planning, stress testing the “missing component in virtualization” GPU. If you are interested you can write to get access to the beta version of LoginVSI.

Here are some screen shots from the session…. watch it to here what Jeroen tells about the upcoming version

Note if you want to get more info on the next version of LoginVSI that supports GPU, write to info@loginvsi.com subject GFX

LoginVSI_gpu_01 LoginVSI_gpu_02

LoginVSI_gpu_03

 

Lakeside

Lakeside Software Monitoring/Assessing NVIDIA GRID

Another groundbreaking product announcement was from Lakeside Software, they are about to release version 7 of Systrack that will support NVIDIA GPU Monitoring/assessing.

Application Graphics Benchmarking

The transformation of an existing software portfolio first begins with the identification of all of the actively used software packages in the environment. The added complication in the case of a project to begin advanced application delivery is the need to understand multiple facets of usage: resource consumption, graphics utilization, frequency of use, user access habits, and mobility needs. Because the state of IT is already so complex it only becomes possible to fully understand and plan with a complete set of descriptive information that really characterizes the unique aspects of every environment. Of particular interest is the ability to first identify applications that have GPU demands, and then begin to segment them into tiers of utilization. SysTrack continually collects information about software packages as they’re used and normalizes all data points for cross platform comparison. One of the key performance parameters that’s identified in this process is a graphical intensity measure (Graphics Index) that provides a way to identify those applications in the portfolio that have higher GPU demands than others. With this critical information it becomes possible to segment the portfolio into groupings based on their requirements for specific resources. By tying a general sense of which applications have peak demand to total length of usage it becomes easier to start developing a portfolio made up of different combinations of usage styles. This includes separating applications that may be used by a small set of the population with intense requirements versus widely used applications with a smaller footprint. Of course, this also allows for much deeper analytics centering on the behaviors of users that is quite important in planning the GPU profiles in use in provisioning. Figure 1 displays this relationship in a bubble chart format, this format groups applications based on their similar characteristics presenting clusters of similar applications in larger bubbles. The vast majority of applications exist in the “low graphics demand – Low Time Active” area in the bottom left, while only a select few have either high graphics demand or high time active.

lakesidesoftware_systrack7-gpu2

SysTrackTracks graphics usage frequency across on physical clients and allows you to group users based on graphics usage & frequency

A natural expansion of this is grouping users into distinct workload types to understand how best to configure the profile types and GPU assignments for users. Once the target applications and users have been characterized and a plan has been developed it’s critical to begin the process of sizing the environment. This includes determining the architecture, sizing the desktops and servers that will be worked with, and identifying resources that will be required to support the needs of the planned deployment.

Resource Modeling & Capacity Planning

NVIDIA Marketplace report from Systrack’sVirtual Machine Planner (VMP) outlines the number of users that fall into different use cases making it easier to forecast how many users per board can be allocated

With a complete portfolio plan it now becomes possible to move into the next phase and start creating a model for what resources will be required for a complete environment. Because each of the users have been fully characterized throughout the assessment data collection interval it’s possible to use SysTrack’s Virtual Machine Planner (VMP) for powerful mathematical analysis to provide deep insight into infrastructure provisioning. The first component of this involves using the profile information above to help develop a plan for what kind of solution will be provided to the end-users. By segmenting the population into different delivery strategies using Citrix FlexCast options as a guideline, a more complete and accurate picture of how the net new environment will operate can be created. An additional benefit of segmentation is the ability to take advantage of grouping by general graphics consumption to identify the number of GPUs required for the environment based on the user density information for each profile type

vgpu-profile

The NVIDIA MarketPlace report from VMP outlines the number of users that fall into the various use cases (e.g. “high” for a designer or higher end power user), making it much easier to forecast how many users per board can be allocated and in turn how many total boards may be needed

lakesidesoftware_systrack7-gpu0

This information creates an easy to use design for a set of user profiles, both for the actual desktop delivery and for the vGPU assignment. By ensuring the best possible analysis of the environment prior to the actual deployment the end-user experience is much simpler to forecast and control. This results in higher end-user satisfaction and a shorter transition time.

User Experience Optimization

After the successful implementation of the solution the environment still requires observation to prevent interruption of service and the potential for productivity impact. The best way to ensure optimal end-user service quality is to have a real-time alerting and analytical engine to collect and report instantly on degradation of any aspect of the systems the users interact with. SysTrack provides this in the form of proactive alerting, detailed system analysis in Resolve, and aggregate trending through Enterprise and Site Visualizer. An even more interesting feature is vScape, a tool designed to examine utilization across multiple virtual machines and correlate resource consumption to concurrency of application utilization. vScape provides real-time updates of all of the application usage across all virtual platforms in an enterprise, including information about what applications are currently demanding GPU resources. It also provides insight into other resource demands as well, such as CPU, memory, and I/O. This can help automate the discovery of co-scheduled or highly concurrent applications to pinpoint the root cause of oversubscription issues much more quickly. It also provides key insight into guest health characteristics with trending to correlate precisely which events may lead to service degradation

lakesidesoftware_systrack7-gpu3

Another key feature introduced in SysTrack version 7.0 is the result of close collaboration with NVIDIA to leverage APIs presented in the guest operating system. This allows the capture of detailed GPU performance metrics to correlate vGPU consumption to end-user service quality. Specifically, with NVIDIA drivers present in the guest OS or on a physical system, the GPU utilization and key metrics (see table 2 for a sample of selected metrics) from the graphics card can be captured and analyzed in the same way as CPU or other system metrics are currently in SysTrack.

lakesidesoftware_systrack7-gpu1

In Systrack 7 after provisioning users in VDI environment the IT admins can monitors performance, which enables to optimize density over time.

This completes the set of KPIs used in SysTrack to calculate the end-user experience score, including categories like resource limitation, network configuration, latency, guest configuration, protocol specific data for ICA, and virtual infrastructure. With a complete set of relevant information the proactive and trending health analysis provided in SysTrack yields a thorough analysis in an easy to understand, quantitative score that summarizes performance on an environmental, group based, or individual system level.

NVIDIA GPU Monitoring/Assessing: (Works with all NVIDIA GPU) Quadro, Kepler, GRID

 

You will be able to look at following parameters:

  • Device ID
  • Power State
  • GPU Usage
  • Frame Buffer Usage
  • Video Usage
  • Bus Usage
  • Memory Usage (Bytes and Percent)
  • # of Apps
  • Temperatures and Fan RPMS

Use this data to accurately plan and size GRID and HDX 3D Pro deployments based on actually observed usage and utilization.

Monitor users post-deployment to provide the best user experience

I recommend reading the whitepaper Lakeside Software have created:
White Paper: SysTrack Delivery Optimization and Planning for NVIDIA GRID and Citrix HDX

 

uberagent

UberAgent 1.8 for Splunk adds GPU performance monitoring

Helge Klein have developed a new version of Splunk that now supports monitoring of GPU, this was a feature request I talked with Helge Klein about in 2013, and I am so happy to see the results what he have done with UberAgent for Splunk, lets dig in what it can do.

uberAgent measures:

  • GPU compute usage per machine
  • GPU memory usage per machine
  • GPU compute usage per process
  • GPU memory usage per process
  • uberAgent shows memory usage separately for shared and dedicated memory (dedicated = on the GPU, shared = main system RAM)
  • uberAgent shows compute usage per GPU engine. The various GPU engines serve different functions, e.g. 2D acceleration, 3D acceleration, video decoding, etc.

uberAgent - process GPU usage uberAgent - single machine GPU usage over time uberAgent - single process GPU usage over time uberAgent - machine GPU usage

For more information visit uberAgent’s website.

My 5 cents

I am very excited to share my findings of some of the things I do in poppelgaard professional services. Feel welcome to contact me at thomas@poppelgaard.com if you are interested in using my professional services and you need help with GPU solutions.

You will see more upcoming blogs from me covering this topic. End User experience, assessments of GPU workload, scaling/sizing, benchmarking, hardware supported, GPU side by side experience, Hypervisor vs Bare metal with a GPU. Watch out for cool things….

Source

Watch the webinar here (YouTube)
Download the presentation here (PDF format)

Lakeside Software
LoginVSI
White Paper: 
SysTrack Delivery Optimization and Planning for NVIDIA GRID and Citrix HDX
UberAgent for Splunk

Citrix XenDesktop HDX3D Pro
Citrix XenApp with GPU Sharing
Citrix XenServer vGPU
NVIDIA GRID
AMD FirePro
VMware vSphere vDGA
VMware vSphere vSGA with NVIDIA GRID

 

Recent Comments

Tobias K

|

Note that on XenServer 7.1, installing VDA 7.13 also works to enable vGPUs running XenDesktop 7.11 even using the newer “NvFBCEnable.exe -enable -noreset” command. Tested with an M60 using various profiles.

Alex

|

So how come version 3.11 doesn’t work for application switching?

John

|

Hello, how to allow SR-IOV for the Intel P580 in a m710x? (unable to find any settings in BIOS)

I get the following error: Dismount-VmHostAssignableDevice : The operation failed.
The device cannot be assigned to a virtual machine as the firmware (BIOS or UEFI) on this host computer system indicates that the device must remain in
contact with the firmware running in the host. The device can only be used in the management operating system. You should contact your OEM to determ
ine if a firmware upgrade is available, or if the PCI Express device can be reconfigured to be independent of the host firmware.

Aza

|

Great article Thomas.

Can you tell more about the difference in user density between M510 and M710x?
Citrix published an article with a LoginVSI test related to the M510 9https://www.citrix.com/blogs/2016/06/30/we-choose-to-go-to-the-moon/), but I’m more interested in the M170x because of its Iris Pro P580.

Olivier

|

Thank you! Thats the solution i needed.