Archive for March, 2016

NVIDIA GRID vGPU 352.83-354.80

Written by Thomas Poppelgaard. Posted in CentOS, Citrix, GRID, GRID 2.0, HDX 3D Pro, Horizon, Linux, Microsoft, NVIDIA, RHEL, Ubuntu, vGPU, VMware, vSphere, Windows 10, Windows 7, Windows 8.1, Windows Server 2008R2, Windows Server 2012R2, XenApp, XenDesktop, XenServer

NVIDIA GRID update - vGPU 352.83-354.80

NVIDIA GRID – vGPU 352.83-354.80

Hi All

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

NVIDIA have released new drivers that supports Windows 10 + fix several bugs.

If you want to test Server 2016 then the new Windows 10 drivers works, I have just tried and they work but they are not “certified” so test this with your own risk, please. Always test this in a test environment and this is not for production environments, unless you want to be early innovators 😉

#HINT – Windows 10 is not supported on Citrix XenServer 6.2 only on Citrix XenServer 6.5.
## HINT – Linux drivers are only available with Tesla M6, M60.

Whats new in vGPU 352.83-354.80

NVIDIA have released a new version of vGPU 352.83-354.80 for NVIDIA GRID 1.0 (K1, K2) GRID 2.0 Tesla M6 and Tesla M60 platform.

Included in this release is

  • NVIDIA GRID Virtual GPU Manager versions 352.83 for Citrix XenServer 6.5 SP1
  • NVIDIA GRID Virtual GPU Manager versions 352.83 for Citrix XenServer 6.2 SP1 with hotfixes XS62ESP1009 and XS62ESP1011
  • NVIDIA GRID Virtual GPU Manager version 352.83 for VMware vSphere 6.0 Hypervisor (ESXi)
  • NVIDIA Windows drivers for vGPU version 354.80
  • NVIDIA Linux drivers for vGPU version 352.83.

 

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

 

Update existing vGPU GRID Manager (Hypervisor)

GRID vGPU Manager 352.83 upgrade for Citrix XenServer 6.2

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

Methodology 1 – the manual way “No GUI”

Upgrading an existing installation of the NVIDIA driver on Citrix XenServer 6.2, use the rpm -U command to upgrade:

If you have NVIDIA GRID K1 / K2
[root@localhost ~]# rpm -Uv NVIDIA-vGPU-kepler-xenserver-6.2-352.83.i386.rpm 
Preparing packages for installation...
If you have NVIDIA GRID TESLA M6 / M60
[root@localhost ~]# rpm -Uv NVIDIA-vGPU-xenserver-6.2-352.83.i386.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

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 352.83, if it is then your host is ready for GPU awesomeness and make your VM rock.

GRID vGPU Manager 352.83 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-6.5-352.83.x86_64.rpm 
Preparing packages for installation...
If you have NVIDIA GRID TESLA M6 / M60
[root@localhost ~]# rpm -Uv NVIDIA-vGPU-xenserver-6.5-352.83.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-6.5-352.83.x86_64-supplemental-pack.tar.bz2 ”

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

“NVIDIA-vGPU-xenserver-6.5-352.83.x86_64-supplemental-pack.tar.bz2 ”

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 352.83, if it is then your host is ready for GPU awesomeness and make your VM rock.

 

GRID vGPU Manager 352.83 for VMware vSphere 6.x

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_352.83-1OEM.600.0.0.2494585.vib
If you have NVIDIA GRID TESLA M6 / M60 select following file:
[root@lesxi ~] esxcli software vib install -v /NVIDIA-vGPU-VMware_ESXi_6.0_Host_Driver_352.83-1OEM.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 352.83, 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.

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 🙂

New driver functionality with 354.80

  • 354.80_grid_win8_win7_64bit_international.exe
  • 354.80_grid_win8_win7_international.exe
  • 354.80_grid_win10_64bit_international.exe
  • 354.80_grid_win10_international.exe
  • NVIDIA-Linux-x86_64-352.83-grid.run

#1 Linux driver is not available with Citrix XenServer 6.2 or 6.5 and ESX for K1/K2. This is only available with GRID vGPU for M60

Linux support in only available with GRID Tesla M6 / M60

GRID vGPU with Linux guest VMs is supported on Tesla M60 and M6, with the following distributions:

  • Red Hat Enterprise Linux 6.6, 7
  • CentOS 6.6, 7
  • Ubuntu 12.04, 14.04 LTS

 

Source

GRID K1/K2 – sources

Download vGPU 352.83-354.80 for XenServer 6.2 for GRID K1 / K2 her

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

Download vGPU 352.83-354.80 for vSphere 6 for GRID K1 / K2 here

 

Tesla M6/M60 – sources

vGPU Grid Manager + Drivers are only available to customers and NVIDIA NPN partners.

Download if you are a NPN partner

Download if you are a GRID 2.0 customer

vGPU 352.83-354.80 for XenServer 6.2 for Tesla M6 / M60

vGPU 352.83-354.80 for XenServer 6.5 for Tesla M6 / M60

vGPU 352.83-354.80 for vSphere 6 for Tesla M6 / M60

For more information about the update from NVIDIA GRID – vGPU 352.83-354.80 contact me.

thomas poppelgaard CTP & MVP

Citrix technology professional – CTP, and Microsoft Most Valuable Professional MVP, Thomas Poppelgaard provides professional services. Write to me on my email thomas@poppelgaard.com or call on my cell +45 53540356

NVIDIA GRID – AppGuide (REVIT)

Written by Thomas Poppelgaard. Posted in Autodesk, GRID, Horizon, NVIDIA, Revit, VMware, vSphere

Hi All.

I want to share this with you all, Luke Wignall and his team from NVIDIA have created some great AppGuides, that helps with understanding how many users can you put on a NVIDIA GRID system with either K1/K2 in a VMware environment. These guides are made together with vendors such as Autodesk. The guides focus on following apps Autodesk Revit.

In this blogpost my goal is to highlight the great work NVIDIA have done creating the scalability app guides and these guides helps you if you want to virtualize Autodesk Revit with NVIDIA GRID and VMware Horizon. The guides are great – cause they give an idea what you would require in a virtualized environment and these can be reused for other virtualized platforms such as Citrix and Microsoft – keep in mind that results might be different. If you would like to get more informations about how the setup is configured and which methodology i recommend you read the AppGuide, you can download it in under source in the end of this article.

The appguides gives a great idea to understand the impact of CPU and how the GPU are giving value.

Autodesk-Logo-2013

revit2015

About Autodesk Revit 2015

Autodesk Revit is Building Information Modeling (BIM) software with features for architectural design, MEP and structural engineering, and construction.  Revit requires a GPU as you rotate, zoom, and interact with drawings. It also creates heavy CPU load as it manages all the elements of a drawing via a database, which means we need high performance storage as well.  The heaviest Revit CPU usage occurs during data-rich operations like file open/save and model updates. As a result both CPU and GPU need to be considered in architecting your vGPU solution.  The size of your drawing file, the concurrency of your users, and the level of interaction with 3D data need to be factored into defining your user groups.

Results Appguide for Autodesk Revit 2015

autodesk-nvidia-vmware

The following are results of our testing, looking for the greatest scalability while still within performance expectations.  Its important to note that your users, your data, and your hardware, will impact these results and you may decide a different level of performance or scalability is required to meet your individual business needs. As the RFO Benchmark does not currently exercise some of Revit’s newest GPU capabilities, and was built to push the limits of dedicated hardware versus the shared resources of VDI, the decision was made to stop testing once the host’s CPU was approaching 100% utilized and test times had climbed past twice what we were finding on the a single physical workstation with dedicated resources.  We then met with the Autodesk Revit team, discussed the results, reviewed the tests in action, and determined with eyes on that this was still within what a typical user would deem acceptable and usable. It’s been well documented that storage performance is key to providing high performance graphics workloads, especially with many users and ever-growing file or model sizes.

FYI – Lower scores are better, representing less time to perform the activity in below tables.

results with Intel Ivy Bridge processors & NVIDIA GRID K2 (K220Q GPU Profile)

revit2015_01

results with Intel Ivy Bridge processors & NVIDIA GRID K2 (K240Q GPU profile)

revit2015_03

results with Haswell processors & NVIDIA GRID K2 (K240Q GPU profile)

revit2015_04

Here are the results when we push for more scalability, bringing the maximum number of VDI guests to 20 and 24.  We have added a 15 second staggered start to emulate synthetic human behavior.

revit2015_05

Revit 2015 users per server

Based on our findings, NVIDIA GRID provides the following performance and scalability metrics for Autodesk Revit 2015; using the lab equipment shown below, using the RFO benchmark, and in working with Autodesk and their emphasis on usability.  Of course, your usage will depend on your models but this is guidance to help guide your implementation.
revit2015-users-pr-server

 

Source

Download the NVIDIA GRID vGPU APPLICATION GUIDE FOR AUTODESK REVIT 2015 ON VMWARE HORIZON here

 

NVIDIA GRID – AppGuide (AutoCAD)

Written by Thomas Poppelgaard. Posted in AutoCAD, Autodesk, Best Practise, Desktop Virtualization, End User Computing, GRID, Horizon, NVIDIA, VMware, vSphere

Hi All.

I want to share this with you all, Luke Wignall and his team from NVIDIA have created some great AppGuides, that helps with understanding how many users can you put on a NVIDIA GRID system with either K1/K2 in a VMware environment. These guides are made together with vendors such as Autodesk. The guides focus on following apps Autodesk AutoCAD.

In this blogpost my goal is to highlight the great work NVIDIA have done creating the scalability app guides and these guides helps you if you want to virtualize Autodesk AutoCAD 2015 with NVIDIA GRID and VMware Horizon. The guides are great – cause they give an idea what you would require in a virtualized environment and these can be reused for other virtualized platforms such as Citrix and Microsoft – keep in mind that results might be different. If you would like to get more informations about how the setup is configured and which methodology i recommend you read the AppGuide, you can download it in under source in the end of this article.

The appguides gives a great idea to understand the impact of CPU and how the GPU are giving value.

Autodesk-Logo-2013

AutoCAD_2016

About AutoCAD 2015

Autodesk AutoCAD is one of the most globally used software platforms for CAD design and documentation.  AutoCAD leverages both CPU and GPU to deliver a high quality user experience, and as a result, there are several things that need to be considered in architecting your GRID vGPU solution: the size of your map data, the concurrency of your users, and the level of interaction with primary 2D and some 3D data.

Results NVIDIA Appguide for Autodesk AutoCAD 2015

autodesk-nvidia-vmware

The following are results of our testing, looking for the greatest scalability while still within performance expectations.  It’s important to note that your users, your data, and your hardware will impact these results, and you may decide a different level of performance or scalability is required to meet your individual business needs. As the Catalyst benchmark does not push AutoCAD’s GPU capabilities, and was built to push the limits of dedicated hardware versus the shared resources of VDI, the decision was made to stop testing once the CPU was approaching 100% utilized and test times had climbed past twice what we were finding on the a single physical workstation with dedicated resources.  We then met with the Autodesk AutoCAD team, discussed the results, reviewed the tests in action, and physically verified that this was still within what a typical user would deem acceptable and usable. When running AutoCAD, anything higher than 20 FPS is awesome, but users generally don’t notice the difference once you exceed 30 FPS.  However, once you drop below 10 FPS, the software is going to feel very sluggish and become unusable by the time you hit 5 FPS. • 20 fps above is good • Below 10fps – sluggish • 5 fps – unusable

  • 32VM : K220Q perform better than K120Q but K120Q is still maintain avg 31FPS. 
  • 40VM : Due to vGPU configuration, need K120Q, avg FPS is 24 FPS
  • 48VM : one you hit 40 VM running, avg FPS is dropping below 10 FPS. 

AutoCAD requires significant CPU resources, so investing in higher CPU speeds and more cores will pay off on performance and scalability.  For medium to large models, K240Q performance might be better for a real use scenario. Since your own situation is different, you must test with your own models to ensure the most accurate results. The chart below shows a comparison between two different sets of VDI guests, one using the K2 and its K220Q profile, the other a K1 and its similarly sized frame buffer profile, the K120Q:

äutocad2015_scaling-01

Notice that as scale, the number of concurrent VM’s increases, there is an obvious drop in performance, and logs will show this is the CPU becoming 100% utilized.  This is a “peak workload” test as all sessions are started in unison, we would expect humans to be more staggered in their workflow.

The next chart shows results when we add that synthetic human behavior, we delay the launching of each session by 5 seconds, this offsets the tests and prevents all tests hitting the same function and impacting the CPU in unison.

äutocad2015_scaling-02
Notice that with delayed launching, or staggered starts, the performance remains high to a much greater scale of concurrent sessions.

Additional findings from this testing:
vCPU –  2 vCPU are sufficient for this benchmark, however increasing model size and complexity will change this. 
• Virtual System Memory – 2 GB is sufficient for this benchmark, however increasing model size and complexity will change this.  
• K220Q/K120Q  : both show capped FPS caused by frame rate limiting (FRL) 
• vGPU has 45 fps frame limiter for performance balancing across multiple VM

AutoCAD 2015 Users per server

Based on the NVIDIA GRID™ Performance Engineering Lab (GRID P.E.L.) findings, NVIDIA GRID provides the following performance and scalability metrics for Autodesk AutoCAD 2015. These metrics are based on tests with the lab equipment shown in the graphic below, using the Cadalyst benchmark, and in working with Autodesk and their emphasis on end user usability.  Of course, your usage will depend on your models and equipment, so this Application Guide is intended to be used as a starting point for your implementation.

autocad-users-per-server

 

Source

Download the NVIDIA GRID vGPU APPLICATION GUIDE FOR AUTODESK AUTOCAD 2015 ON VMWARE HORIZON here

 

NVIDIA GRID – AppGuide (ArcGIS Pro)

Written by Thomas Poppelgaard. Posted in Best Practise, End User Computing, ESRI, GRID, Horizon, NVIDIA, VMware

Hi All.

I want to share this with you all, Luke Wignall and his team from NVIDIA have created some great AppGuides, that helps with understanding how many users can you put on a NVIDIA GRID system with a K2 in a VMware environment. These guides are made together with vendors such as ESRI. The guides focus on following apps ESRI ArcGIS Pro.

In this blogpost my goal is to highlight the great work NVIDIA have done creating the scalability app guides and these guides helps you if you want to virtualize ESRI Pro with NVIDIA GRID and VMware Horizon. The guides are great – cause they give an idea what you would require in a virtualized environment and these can be reused for other virtualized platforms such as Citrix and Microsoft – keep in mind that results might be different. If you would like to get more informations about how the setup is configured and which methodology i recommend you read the AppGuide, you can download it in under source in the end of this article.

The appguides gives a great idea to understand the impact of CPU and how the GPU are giving value.

esri

arcgis

About ESRI ArcGIS Pro

ESRI ArcGIS Pro 1.0 is a  Geographic Information Systems (GIS) application for mapping, visualizing, editing, and analyzing spatial data.  Esri recommends a GPU for best end user experience, but as ArcGIS Pro 1.0 also generates heavy CPU load, this also needs to be considered in architecting your vGPU solution.  The size of your map data, the concurrency of your users, and the level of interaction with 3D data all need to be considered when defining your user groups.

Results NVIDIA Appguide for ESRI ArcGis Pro

esri-nvidia-vmware

The following are the full results of our testing.  The baseline was the 45 second draw time sum – anything greater than that value represented a worsening UX while less would be a better UX.  Looking for both performance and scalability, we tested looking for the greatest number of virtual desktops, and therefore the greatest scalability, while still within performance expectations and the threshold of 45 seconds.  It’s important to note that your users, your data, and your hardware will impact these results and you may decide a different level of performance or scalability is required to meet your individual business needs.   Tests were also run to look for potential NUMA issues that can negatively impact performance.  This is where the physical GPU and its PCI-e channels are tied to one physical CPU, while the virtual desktop is running on the other physical CPU, so communication with the physical GPU has to move over the QPI between the two physical CPUs.  This creates a bottleneck and can cause performance issues.  However, in our testing, the application is sufficiently CPU bound that NUMA affinity made little difference. The results in the table below show the decrease in performance as we increased vCPU counts, and then the increase in scalability with synthetic human behavior (think time):

esri-arcgis-users-pr-server-2

ArcGIS – users per server

Based on NVIDIA Performance Engineering Lab findings, NVIDIA GRID provides the following performance and scalability metrics for Esri ArcGIS 3D Pro 1.0. These metrics are based on tests with the lab equipment shown in the graphic below, using the Esri API based “heavy 3D” benchmark and in working with Esri to determine acceptable performance.  Of course, your usage will depend on your models, but this is guidance to help guide your implementation.

esri-arcgis-users-pr-server

Source

Download the NVIDIA GRID vGPU APPLICATION GUIDE FOR ESRI ARCGIS PRO 1.0 – 3D ON VMWARE HORIZON here

 

Recent Comments

MarilynToush

|

This message is posted here using XRumer + XEvil 3.0
XEvil 3.0 is a revolutionary application that can bypass almost any anti-botnet protection.
Captcha Recognition Google, Facebook, Yandex, VKontakte, Captcha Com and over 8.4 million other types!
You read this – it means it works! 😉
Details on the official website of XEvil.Net, there is a free demo version.

Server 2016 Support

|

Wonderful website. A lot of useful info here. I am sending it to a few friends ans also sharing in delicious. And obviously, thanks for your sweat!

Server 2016 Support

Thomas Poppelgaard

|

Thanks Jose, I have corrected the map plus updated the map so 29 datacenters are now alive.

Jose Brenes

|

Your map of Australia is wrong. You have put the state of Victoria inside the West Australian state.

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.