NVIDIA GRID Virtual Workstation, CATIA, M60, M6 Tesla, K2, K240Q, K260Q, K280Q, M60 1Q, M60 2Q, M60 4Q GRID Scalability Guide

NVIDIA GRID
DASSAULT CATIA V5/V6
SCALABILITY GUIDE
NVIDIA Performance Engineering Labs
PerfEngDoc-SG-DSC01v1
March 2016
HOW MANY USERS CAN I GET ON A SERVER?
The purpose of this guide is to give a detailed analysis of how many users organizations
can expect to get per server based on the application being used and the types of users
they have in their environment. The data for this analysis is sourced from a combination
of application recommendations from the independent software vendor (ISV), along
with customer data from actual deployments. As with any deployment, your usage will
depend on your models and equipment, so this Scalability Guide is best used as a
starting point for your implementation.
DASSAULT SYSTÈMES CATIA
Dassault Systèmes CATIA is one of the most globally used CAD (Computer Aided
Design) applications, being used for almost all aircraft designs in the world and over
80% of the automotive market. As a result, the use of CATIA is pervasive throughout the
supply chains supporting these industries. CATIA leverages both CPU and GPU to
deliver a high quality user experience so there are several things that need to be
considered in architecting your NVIDIA GRID Virtual Workstation (vWS) solution.
These include; the size of your data, the concurrency of your users, and the level of
interaction with 3D data.
Recommended CATIA Virtual System Requirements
Working with virtualization partners like Citrix and VMware, and alongside our shared
customers with tested and/or production environments, we are recommending the
following system requirements for deploying CATIA in a virtual environment. Again,
this is a starting point and your requirements will depend on your users and data sets,
testing with those users will provide the best guidance:
Recommended Virtualization Platforms
Citrix XenServer v6.5 SP1 or later
Virtualization
Software
Citrix XenDesktop v7.8 or later
VMware vSphere 6 or later
VMware Horizon 7 or later
Virtual Machine
Operating System
Microsoft Windows 7 SP1 64-bit: Enterprise, Ultimate, or Professional
Microsoft Windows 8.1 64-bit: Enterprise, or Professional
Host Server
Recommendation
Minimum
Recommended
Advanced
CPU
2.3 GHz+ Intel® Xeon
E5 v3 or greater
2.6 GHz+ Intel® Xeon
E5 v3 or greater
2.6 GHz+ Intel® Xeon E5
v3 or greater
Memory
196 GB
320 GB
320 GB
10 G
10 G or greater
~750+ IOPS Per User
~1000+ IOPS Per User
Networking
Storage
1 GB minimum
10 G recommended
~500+ IOPS Per User
2
www.nvidia.com
NVIDIA Tesla
M60/M6 or greater
NVIDIA Tesla
M60/M6 or greater
NVIDIA Tesla M60/M6
or greater
NVIDIA GRID K2 or
greater
NVIDIA GRID K2 or
greater
NVIDIA GRID K2 or
greater
Virtual Machine
Settings
Minimum
Recommended
3D & Large Datasets
Memory
8 GB RAM
16 GB RAM
24 GB RAM or greater
vCPUs
4 vCPUs
4+ vCPUs
6+ vCPUs
Disk Space
100 GB free disk space
100 GB free disk space
100 GB free disk space
NVIDIA GRID
Quadro vWorkstation
License
NVIDIA GRID
Quadro vWorkstation
License
NVIDIA GRID Quadro
vWorkstation License
K240Q / M60-1Q
profile or greater
K260Q / M60-2Q
profile or greater
GPU
Virtual Graphics
License & Profile
K280Q / M60-4Q profile
or greater
CATIA SCALABILITY BASED ON CUSTOMER DATA
Based on testing and production results from several customers in the manufacturing
space, here are the expected user densities for the different user types on NVIDIA GRID
environments with GRID software and NVIDIA Tesla® M60 or GRID K2 cards. Please
note that these results are what these customers determined were appropriate for their
needs and provide insight into what one might expect. Results vary based on many
variables and doing your own testing will determine your specific scalability.
TECH TIP! It is important to note that mixed user types have produced the best density for
some customers, for example placing 1 large and 7 mediums user types per host.
CATIA User
Classifications
Application(s)
Small/Standard
Dassault Systèmes
CATIA
Description
Users per
server
View-Only or Full application
Accessing individual parts or small
assemblies
8-16
Manufacturing shop floor
Medium
Dassault Systèmes
CATIA
View-Only or Full application
Accessing medium assemblies
8-16
Manufacturing shop floor
Large/Power
User
Dassault Systèmes
CATIA
Full Application
Analyst
Dassault Systèmes
CATIA
Full Application
Accessing large assemblies or full
model
CUDA/OpenCL
4-8
4
Analysis of/or accessing large
assemblies or full model
3
www.nvidia.com
TYPICAL CATIA VIRTUAL WORKSTATION BUILDS
Dassault Systèmes and NVIDIA are developing recommendations and certifications for
architecting your virtualized CATIA deployment. The following are actual builds from
current customers, in production or POC use today, and can provide a starting point for
building your own environment. Your own tests with your own models will determine
if these recommendations meet your specific needs.
Automotive Manufacturing:
Three automotive manufacturers provided data for CATIA on GRID software with K2
and M6/M60 GPUs. This solution is being used to centralize data and compute into
global regions, improving productivity, collaboration, supply chain (see below)
integration, and of course security.
Two automotive parts manufacturers provided builds and user profiles that they
architect for. These users are either located on a metro area campus setting, or in satellite
offices. Having to work with multiple automotive manufacturers requires parts
suppliers to work with multiple CATIA versions, and the easiest way to achieve this is
with virtual workstations. Virtualization also allows the parts suppliers to centralize
their data, collaborate with ease, distribute their workforce, and retain intellectual
property.
Resulting User Builds:
User Classification Matrix
Customer User
Classifications
CATIA Build Spec
STD CAD Desktop
ADV CAD Desktop
ADV+ CAD Desktop
4-6 vCPUs
6 vCPUs
8-16 vCPUs
16GB RAM
16GB RAM
16GB RAM
1GB Frame Buffer
2GB Frame Buffer
4GB Frame Buffer
(K240Q / M60-1Q)
(K260Q / M60-2Q)
(K280Q / M60-4Q)
Exported model
size
150-200 MB
500-600 MB
Total triangles
~100,000
~1,600,00
122ms/52ms
937ms/386ms
Visualization
time (CPU vs vGPU
Users per host (dual CPU,
2 GRID K2’s per host)
NVIDIA GRID Software
Users per host (dual CPU,
2 Tesla M60’s per host)
16
8
4
Virtual Workstation (vWS) Edition
32
16
4+
(CUDA / OpenCL
req.)
4
www.nvidia.com
Aerospace manufacturing:
This aerospace manufacturer has four different user profiles that they architect for. They
have run production users of CATIA workloads on both GRID K2 GPUs as well as
GRID software and with Tesla M60 GPUs. These users were located in satellite offices,
connected via enterprise class WAN links where a 70% productivity improvement was
reported thanks to file proximity/transfer time savings alone, to those based on the LAN
where even then a 30% productivity gain was reported. With Frame Rate Limiting
(FRL) turned on, typical and appropriate for production environments, they saw no
increase in Frame Per Second (FPS) but were able to double the number of users per host
with no reported degradation of end user performance.
Resulting User Builds:
User Classification Matrix
Customer User
Classifications
Small/Standard
CATIA Build
Spec
Users per host
(dual CPU, 2
GRID K2 per
host)
NVIDIA GRID
Software
Users per host
(dual CPU, 2
GRID M60 per
host)
Large/Power User
Jumbo/Analyst
2vCPU
16GB RAM
1GB Frame Buffer
(K240Q / M60-1Q)
Medium/
Moderate
4vCPU
32GB RAM
1GB Frame Buffer
(K240Q / M60-1Q)
4vCPU
64GB RAM
1GB Frame Buffer
(K240Q / M60-1Q)
4vCPU
128GB RAM
1GB Frame Buffer
(K240Q / M60-1Q)
16
16
8
4
Virtual Workstation (vWS) Edition
32
32
16
8
We will continue to collect field data from customers and partners to further refine this
data. If you would like to share your data please reach out to your local NVIDIA team
and let us know.
ANALYSIS SUMMARY
Dassault Systèmes CATIA requires significant CPU resources, so investing in higher
CPU speeds and more cores will pay off on performance and scalability. Smaller
workloads can leverage the 1GB vGPU profile, but for medium to large models, 2GB or
greater vGPU profile performance generally produce better results. Since each situation
is different, you must test with your own models to ensure the most accurate results.
To test NVIDIA GRID in your environment you can choose to get started with a certified
NVIDIA partner or start a POC yourself with a certified server and our 90-day
evaluation licenses.
5
www.nvidia.com
Important things to remember during your POC:
1. Define “acceptable” user experience: Defining user experience (UX) requires
careful examination of user and application interaction. This can be obvious, like
the rendering time for an image to appear, or smoothly panning across that
image. It can also be less obvious, like the ability to smoothly scroll down a page
or the “snappy” reaction for a menu to appear after a right click. Ask users to
report metrics, and to judge specific activities or functions using finite scales (e.g.
1-5, 5 being best), to avoid generic feedback.
2. Compare real world workloads: In virtual environments, time-slicing of
resources allows users to get the same level of performance even when sharing
resources. This is due to user “think time” which includes any time the user is
not actually interacting with the application, or when not using the application
or even sitting at their workstation. Add up all the time away from the
application (meetings, lunch, out of office, etc.) and one could expect to get even
more benefits from shared resources. These benefits equate to more resources for
the user’s session and typically a more responsive application, thus a betterperceived experience by the end user, as opposed to peak workload benchmarks
with inhuman like uninterrupted work.
3. Test with real users: It’s important to actually look at the application running be
sure that the experience is enjoyable for users. That being said, it’s also
important to maintain perspective, especially if you are not a regular user of
applications like Dassault Systèmes CATIA. While a data center admin
deploying CATIA in a virtual environment might view a testing desktop and
think the experience is slow or sluggish, a user who works with it daily might
find it normal. The feedback from an actual 3D designer using the application in
a virtual desktop is the ultimate test of success. Add in the point above about
real world workloads and you see why real users are the most accurate means of
testing.
6
www.nvidia.com
Notice
The information provided in this specification is believed to be accurate and reliable as of the date provided. However, NVIDIA
Corporation (“NVIDIA”) does not give any representations or warranties, expressed or implied, as to the accuracy or completeness
of such information. NVIDIA shall have no liability for the consequences or use of such information or for any infringement of
patents or other rights of third parties that may result from its use. This publication supersedes and replaces all other
specifications for the product that may have been previously supplied.
NVIDIA reserves the right to make corrections, modifications, enhancements, improvements, and other changes to this
specification, at any time and/or to discontinue any product or service without notice. Customer should obtain the latest relevant
specification before placing orders and should verify that such information is current and complete.
NVIDIA products are sold subject to the NVIDIA standard terms and conditions of sale supplied at the time of order
acknowledgement, unless otherwise agreed in an individual sales agreement signed by authorized representatives of NVIDIA and
customer. NVIDIA hereby expressly objects to applying any customer general terms and conditions with regard to the purchase
of the NVIDIA product referenced in this specification.
NVIDIA products are not designed, authorized or warranted to be suitable for use in medical, military, aircraft, space or life
support equipment, nor in applications where failure or malfunction of the NVIDIA product can reasonably be expected to result
in personal injury, death or property or environmental damage. NVIDIA accepts no liability for inclusion and/or use of NVIDIA
products in such equipment or applications and therefore such inclusion and/or use is at customer’s own risk.
NVIDIA makes no representation or warranty that products based on these specifications will be suitable for any specified use
without further testing or modification. Testing of all parameters of each product is not necessarily performed by NVIDIA. It is
customer’s sole responsibility to ensure the product is suitable and fit for the application planned by customer and to do the
necessary testing for the application in order to avoid a default of the application or the product. Weaknesses in customer’s
product designs may affect the quality and reliability of the NVIDIA product and may result in additional or different conditions
and/or requirements beyond those contained in this specification. NVIDIA does not accept any liability related to any default,
damage, costs or problem which may be based on or attributable to: (i) the use of the NVIDIA product in any manner that is
contrary to this specification, or (ii) customer product designs.
No license, either expressed or implied, is granted under any NVIDIA patent right, copyright, or other NVIDIA intellectual property
right under this specification. Information published by NVIDIA regarding third-party products or services does not constitute a
license from NVIDIA to use such products or services or a warranty or endorsement thereof. Use of such information may require
a license from a third party under the patents or other intellectual property rights of the third party, or a license from NVIDIA
under the patents or other intellectual property rights of NVIDIA. Reproduction of information in this specification is permissible
only if reproduction is approved by NVIDIA in writing, is reproduced without alteration, and is accompanied by all associated
conditions, limitations, and notices.
ALL NVIDIA DESIGN SPECIFICATIONS, REFERENCE BOARDS, FILES, DRAWINGS, DIAGNOSTICS, LISTS, AND OTHER DOCUMENTS
(TOGETHER AND SEPARATELY, “MATERIALS”) ARE BEING PROVIDED “AS IS.” NVIDIA MAKES NO WARRANTIES, EXPRESSED, IMPLIED,
STATUTORY, OR OTHERWISE WITH RESPECT TO THE MATERIALS, AND EXPRESSLY DISCLAIMS ALL IMPLIED WARRANTIES OF
NONINFRINGEMENT, MERCHANTABILITY, AND FITNESS FOR A PARTICULAR PURPOSE. Notwithstanding any damages that customer
might incur for any reason whatsoever, NVIDIA’s aggregate and cumulative liability towards customer for the products described
herein shall be limited in accordance with the NVIDIA terms and conditions of sale for the product.
VESA DisplayPort
DisplayPort and DisplayPort Compliance Logo, DisplayPort Compliance Logo for Dual-mode Sources, and DisplayPort Compliance
Logo for Active Cables are trademarks owned by the Video Electronics Standards Association in the United States and other
countries.
HDMI
HDMI, the HDMI logo, and High-Definition Multimedia Interface are trademarks or registered trademarks of HDMI Licensing LLC.
ROVI Compliance Statement
NVIDIA Products that support Rovi Corporation’s Revision 7.1.L1 Anti-Copy Process (ACP) encoding technology can only be sold
or distributed to buyers with a valid and existing authorization from ROVI to purchase and incorporate the device into buyer’s
products.
This device is protected by U.S. patent numbers 6,516,132; 5,583,936; 6,836,549; 7,050,698; and 7,492,896 and other
intellectual property rights. The use of ROVI Corporation's copy protection technology in the device must be authorized by ROVI
Corporation and is intended for home and other limited pay-per-view uses only, unless otherwise authorized in writing by ROVI
Corporation. Reverse engineering or disassembly is prohibited.
OpenCL
OpenCL is a trademark of Apple Inc. used under license to the Khronos Group Inc.
Trademarks
NVIDIA, the NVIDIA logo, NVIDIA GRID, and NVIDIA GRID vGPU are trademarks and/or registered trademarks of NVIDIA Corporation
in the U.S. and other countries. Other company and product names may be trademarks of the respective companies with which
they are associated.
Copyright
© 2016 NVIDIA Corporation. All rights reserved.
7
www.nvidia.com