NOW: With advanced technology, you can do even more with GPUs: Engineers use GPUs to design everything from golf clubs to luxury cars to jumbo jets. Researchers use GPUs to model complex systems, from the human body to the universe. GPUs create stunning visual effects for award-winning movies, media and games. Scientists use GPUs to advance medical science, prevent pandemics, and discover new drugs. WHAT IS A GPU: A GPU, or graphics processing unit, is a highly efficient parallel processor that excels at tackling large amounts of data simultaneously. The beauty of the 3D world comes to life and complex calculations are exponentially faster. DID YOU KNOW? > More than 1 billion GPUs have been shipped to date. > NVIDIA has 5,000+ patents granted or pending. > GPUs can accelerate applications by as much as 100x. > NVIDIA GPUs power 50 of the world’s top supercomputers. > 100% of nominees for Best Visual Effects Oscar were powered by NVIDIA GPUs (2010, 2011, 2012, and 2013). > NVIDIA received a Technology and Engineering Emmy Award for its innovation in graphics processor development. HOW GPUs WORK: GPUs and CPUs work together to process application workloads. The CPU handles simple serial tasks while the GPU handles increasingly more visual and compute-intensive workloads. The result is a transformation of everyday PCs into professional workstations and supercomputers. A CPU has multiple cores A GPU has thousands of cores THEN: In 1999, GPUs were used to develop graphics for PC games. NVIDIA ® PROFESSIONAL SOLUTIONS SALES TOOLKIT INSPIRING CUSTOMERS TO DO THEIR BEST WORK. FASTER. POWERFUL | EFFICIENT | EXTRAORDINARY Imagine a processor that offers advanced design capabilities for entertainment and commerce, while enabling scientists to tackle the world’s most challenging problems from climate research to quantum physics. The GPU makes all of this possible. NVIDIA customers want to do things better, faster and at higher capacity than ever before. Help your prospective customers realize their ambitions with the right professional solution from NVIDIA. EVOLUTION OF THE GPU WHAT IS A GPU?
Simple guide to understanding customers needs and positioning the best Nvidia solution, This is an easy-to-use Sales Guide that we provide to our partners.
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
NOW: With advanced technology, you can do even more with GPUs:
Engineers use GPUs to design everything from golf clubs to luxury cars to jumbo jets.
Researchers use GPUs to model complex systems, from the human body to the universe.
GPUs create stunning visual effects for award-winning movies, media and games.
Scientists use GPUs to advance medical science, prevent pandemics, and discover new drugs.
WHAT IS A GPU: A GPU, or graphics processingunit, is a highly efficient parallel processor that excels at tackling large amounts of data simultaneously. The beauty of the 3D world comes to life and complex calculations are exponentially faster.
DID YOU KNOW? > More than 1 billion GPUs have been shipped to date.
> NVIDIA has 5,000+ patents granted or pending.
> GPUs can accelerate applications by as much as 100x.
> NVIDIA GPUs power 50 of the world’s top supercomputers.
> 100% of nominees for Best Visual Effects Oscar were powered by NVIDIA GPUs (2010, 2011, 2012, and 2013).
> NVIDIA received a Technology and Engineering Emmy Award for its innovation in graphics processor development.
HOW GPUs WORK: GPUs and CPUs work together to process application workloads. The CPU handles simple serial tasks while the GPU handles increasingly more visual and compute-intensive workloads.The result is a transformation of everyday PCs into professional workstations and supercomputers.
A CPU hasmultiple cores
A GPU has thousands of cores
THEN: In 1999, GPUs were used to develop graphics for PC games.
NVIDIA® PROFESSIONAL SOLUTIONS SALES TOOLKITINSPIRINg CUSTOmERS TO DO ThEIR BEST WORK. FASTER.
Powerful | effICIeNT | eXTrAorDINArYImagine a processor that offers advanced design capabilities for entertainment and commerce, while enabling scientists to tackle the world’s most challenging problems from climate research to quantum physics. The GPu makes all of this possible. NVIDIA customers want to do things better, faster and at higher capacity than ever before. Help your prospective customers realize their ambitions with the right professional solution from NVIDIA.
eVoluTIoN of THe GPu
WHAT IS A GPU?
NVIDIA offers a suite of professional solutions that enable designers, manufacturers, scientists and researchers to do their work smarter, faster, with greater reliability and with energy efficiency. Help your customer choose the right GPU with the following conversation starter.
wHY DoeS Your CuSToMer NeeD A GPu?
I worK wITH BuSINeSS
APPlICATIoNS THAT Are NoT
HIGHlY GrAPHICS INTeNSIVe BuT I
NeeD To relIABlY DrIVe MulTIPle
DISPlAYS
I worK wITH APPlICATIoNS THAT
Are 3D GrAPHICS INTeNSIVe AND I NeeD MoBIlITY
I worK wITH APPlICATIoNS THAT
Are 3D GrAPHICS INTeNSIVe AND I uSe A DeSKToP worKSTATIoN
I worK wITH APPlICATIoNS THAT HAVe 3D GrAPHICS BuT
AlSo reQuIre HIGH CoMPuTATIoNAl
HorSePower IN MY DeSKToP worKSTATIoN
I worK wITH HIGH-PerforMANCe
CoMPuTATIoNAl APPlICATIoNS
(HPC) oN A DeSKToP
worKSTATIoN
I worK wITH HIGH-PerforMANCe
CoMPuTATIoNAl APPlICATIoNS
(HPC) IN A SerVer eNVIroNMeNT
DoeS Your worK INVolVe CreATING or VISuAlIzING DATA AND IMAGeS, or HeAVY CoMPuTATIoN?
QueSTIoN YeS
See NVIDIA TeSlA section for more information
See NVIDIA QuADro and NVIDIA TeSlA sections for more information
See NVIDIA QuADro section for more information
See NVIDIA QuADro section for more information
See NVIDIA NVS section for more information
reCoMMeND NVIDIA TeSlA®
The high-performance solution
> NVIDIA Tesla parallel processors can handle the largest datasets in science and engineering applications.
reCoMMeND NVIDIA MAXIMuS™Visualization and computation solution
> NVIDIA Maximus combines the visualization capability of Quadro with the computing power of Tesla in a single workstation.
reCoMMeND NVIDIA NVS™The multi-display solution
> NVIDIA NVS enhances productivity in multi desktop environments.
reCoMMeND NVIDIA QuADro®
Powerful visualization for MOBILE
> NVIDIA Quadro Mobile offers the same performance as Quadro Desktop solutions in a portable form factor.
reCoMMeND NVIDIA QuADro®
Powerful visualization for DESKTOP
> NVIDIA Quadro is recommended, tested, and certified on 200+ professional applications.
Great! NVIDIA Professional Solutions are perfect for you!
NVIDIA® NVS™
mULTI‑DISPLAy SOLUTION
NVIDIA NVS is an enterprise business graphics solution that enhances productivity in multi-desktop environments. Key benefits include:
> Flicker-free viewing for two or more displays
> Compatible across multiple display and system types
> Built-in NVIDIA technologies to easily manage your display real estate*
- NVIDIA® nView® desktop management software that helps increase productivity
- NVIDIA Mosaic technology that seamlessly spans your entire desktop across multiple displays
> Flexibility to expand the number of displays without replacing your system**
fINANCIAl INSTITuTIoNS> Trading Floors > Insurance and Banking Centers
DIGITAl SIGNAGe> Airport Signage > Restaurants and Shopping Malls
ProCeSS CoNTrol> Government Institutions > Control and Command Centers
BuSINeSS eNVIroNMeNTS> Call Centers > Hospitals
* For more information on nView and Mosaic technologies please visit www . nvidia . com / nview and www . nvidia . com / mosaic | ** Not available with integrated graphics options NVIDIA
NVS
NVS 300
NVS 310
NVS 420
NVS 450
NVS 510
Consumer Graphics Card
Integrated Graphics
ADVANCeD DISPlAY CAPABIlITY Max Number of Displays 2 2 4 4 4 2-4 2
* NOTE: Integrated graphics serviceability requires you to replace system.
1 Requires a DMS59 to DVI/VGA/DisplayPort cable adaptor | 2 Requires a VHDCI to DVI/DisplayPort based cable adaptor 3 NVS 510 supports a maximum of 4 simultaneous displays in Win7, Win8 & Linux OS’s only (maximum of 2 displays in Windows XP)
Enterprise-Management Tools Enterprise tools for remote management of graphics and display settings. Driver-installer tools.
reCoMMeNDeD SoluTIoN for CuSToMerS wHo NeeD...
> Graphics to drive four displays in small form-factor or full-height systems
> Best performance and low power
> Large dedicated frame buffer
> Graphics to drive four displays in full-height desktops or workstations
> Low power consumption and fan-less solution
> Graphics to drive four displays in small form-factor or full-height systems
> Low power consumption
> Graphics to drive one or two displays via VGA, DVI, or DisplayPort
> Lowest power and fan-less solution
> Graphics to drive one or two displays via DisplayPort
> Balanced power and performance
wHICH NVIDIA® NVS™ GPu DoeS Your CuSToMer NeeD?
3
NVIDIA
NVS
HelP Your CuSToMerS CoNNeCT THeIr DISPlAYS
For NVS 510, use included mini DP-to-DP cable with standard DisplayPort cable/cable adaptor to connect to any display type. Alternately users can also directly connecting using mini DisplayPort cables/adaptors.
1 Requires a DMS59 to DVI/VGA/DisplayPort cable adaptor2 Requires a VHDCI to DVI/DisplayPort-based cable adaptor3 Requires a DisplayPort to SL-DVI/VGA/HDMI cable adaptor4 Requires DisplayPort to Dual Link DVI dongle
Number of Displays
Max resolution of Display Preferred Connectors/Cable 1-2 3-4 5-6 6-8
NVIDIA QuADro family of products, based on the Kepler architecture GPUs are recommended, tested, and certified on more than 200 professional applications across many industries, including manufacturing, entertainment, sciences, and energy. Professionals trust them to deliver the best possible experience with applications.
NVIDIA QuADro MoBIle GPUs offer comparable performance in a mobile solution.
NVIDIA MAXIMuS second generation technology based on the Kepler architecture combines two or more NVIDIA processors to accelerate graphics and computation at the same time. In the Maximus configuration:
> The NVIDIA Tesla® GPU performs the heavy lifting of photorealistic rendering, simulation, or computation.
> The NVIDIA Quadro GPU powers rich, full-performance, interactive design.
> The CPU is freed up for the tasks it performs best.
NVIDIA QuADro
NVIDIA MAXIMuS™ reCIPe*
ISV certified drivers supporting Maximus: 310 or newer
Visual Computing GPu:
Quadro K5000Quadro K4000Quadro K2000Quadro K600
Parallel Computing GPu:
Tesla K20c(one or more) +
effICIeNT MulTI-TASKING
HIGHer PerforMANCe
ISV CerTIfIeD
✓Design AND simulate ✓Faster simulation ✓Tesla and Quadro board certifications
✓ Design AND render ✓Faster rendering ✓Global OEM system certifications
✓ Seamless experience ✓100% interactive design performance while rendering or simulating
✓All leading CAD and CAE ISVs
MAXIMuS BeNefITS
DeSIGN reNDer/
SIMulATe GPu
CPu
*Kepler Family NVIDIA
QUADRO
reCoMMeNDeD NVIDIA ProfeSSIoNAl SoluTIoNgOOD BETTER BEST
> Occasional use of application > Regular use of application > Pushing the application to its limits
> Small objects and assemblies > Average to large objects and assemblies > Extremely large objects and assemblies
> About 10% of the users > Using all application features > Using all application features
ADAPTorSDP→DVI Up to 1920x1200 display resolution - use a DP to DVI-SL (single link) adapter Above 1920x1200 display resolution - use a DP to DVI-DL (dual link) adapter
DP→HDMI
DP→VGA
DVI→VGA
DVI→HDMI
mDP→DP
mDP→DVI-SL
*Use a mDP→DP adaptor to connect the DP→HDMI or DP→VGA adaptor.
*
KeY ✓= Connect directly no adapter required.
Example: “DP→DVI” = Requires a DisplayPort to DVI adapter. “mDP→HDMI” = Requires a mini DisplayPort to HDMI adapter.
If you want to connect a display with the following cable
NVIDIA® TESLA®
hIgh‑PERFORmANCE COmPUTINg
NVIDIA TeSlA GPU Accelerators consist of thousands of cores to quickly compute large workloads, which make them perfect for accelerating scientific and engineering applications that often run in server environments.
wHICH CuSToMerS wIll BeNefIT froM TeSlA?
KeY BeNefITS KeY BeNefITS
wHo To TAlK To
IT MANAGerS reSeArCHerS & APPlICATIoN uSerS
GPUs deliver high performance and capacity across many users, enabling:
> Maximum performance in a smaller footprint
> More compute power for the budget: Tesla GPUs accelerate a wide variety of applications and are optimized to provide maximum performance per watt.
> energy efficiency: GPUs are inherently energy efficient, making them ideal for increasing performance in data centers and running computationally intensive science and engineering applications.
GPUs accelerate computationally intensive applications. By supporting more and faster iterations, GPUs enable:
> lower cost of innovation
> faster innovation: For example, molecular dynamics uses complex computer simulations to study physical interactions of atoms and molecules, which can speed up innovation in drug discovery.
> Higher quality results
> faster time to market
> fewer expensive lab experiments (can now be replaced by GPU-powered simulations)
> Powerful simulation: For example, Princeton University researchers are using GPUs to simulate insect behavior and save crops from destructive locust swarms.
Medical & life Science Instruments> Imaging & Analysis
Media & entertainment> Rendering> Encoding & Digital Distribution> Video Editing> Color Correction> Simulation
INDuSTrIeS
NVIDIA
TESLA
ToP GPu-ACCelerATeD APPlICATIoNS
DeVeloPer ToolS for ProPrIeTArY APPlICATIoNSCustomers with their own proprietary applications can also accelerate them with GPUs. Direct your customer to NVIDIA’s extensive suite of developer tools at: www.nvidia.com/parallelprogramming
TAKe A GPu for A free TeST DrIVe!Your customers can easily experience the performance of NVIDIA GPU Accelerators by taking a free GPU Test Drive:
> Step 1: Register at www.nvidia.com/gputestdrive> Step 2: Log into a remote server and try the latest GPU-accelerated applications, or load the proprietary code.
Over 200 third-party applications have already been GPU-accelerated. Here are some of the most popular ones. For the full list, visit www.nvidia.com/appscatalog
APPlICATIoNPerforMANCe IMProVeMeNT VS. CPu-oNlY SYSTeMS
NVIDIA®Tesla® GPU Accelerators are ideal for optimizing performance in small and medium server clusters, as well as the world’s most powerful supercomputers.
TerM DeSCrIPTIoNAssembly/Model An assembly is made up of a part of a physical product that contains multiple 3D sub-components
or models.
CAD Computer-aided design (CAD), also known as computer-aided drafting (CAD) or computer-aided design and drafting (CADD), is the use of computer systems to assist in the creation, modification, analysis, or optimization of a design. Computer-aided drafting describes the process of creating a technical drawing with the use of computer software. CAD software is used to increase the productivity of the designer, improve the quality of design, improve communications through documentation, and to create a database for manufacturing.
CAe Computer-aided-engineering (CAE) is the broad usage of computer software to aid in engineering tasks. For example, CAE systems are able to simulate a design under a variety of conditions to see if it actually works.
Computational Simulations Computer simulation is the attempt to simulate through a computer model an abstract model of a particular system, such as molecules and materials.
CuDA NVIDIA® CUDA™ is a parallel computing platform and programming model invented by NVIDIA. It enables dramatic increases in computing performance by harnessing the power of the graphics processing unit (GPU).
DCC Digital content creation is a term that refers to the creation and modification of digital content, such as animation, audio, graphics, images and video, as part of the production process before presentation in its final form. This large field encompasses many segments such as 3D graphics, audio editing, compositing, authoring, etc.
Directives openACC Allows programmers to provide simple hints, known as directives, to the compiler to accelerate code. With OpenACC directives, the code remains portable across many-core GPUs and multi-core CPUs.
fermi NVIDIA® Fermi is the code name for the NVIDIA GPU architecture that launched in 2009. NVIDIA products built on the Fermi architecture have names such as Quadro 2000 or Tesla M2090, for example.
frame Buffer Frame buffer is dedicated memory on the graphics card similar to that which is used to store model geometry, textures, frames of video, etc.
GPu Graphics processing unit with hundreds of compute cores, initially used as graphic processors.
GPu Accelerator Now that GPUs have become programmable and computationally powerful, they are also used to accelerate hundreds of scientific applications.
GPu Computing or GPu Accelerated Computing
Use of a GPU (graphics processing unit) together with a CPU to accelerate general-purpose scientific and engineering applications.
GPu ray Tracing Use of the GPU to greatly accelerate ray trace rendering in solutions such as Iray, VRay-RT, Octane, FinalRender, etc.
GPu Test Drive A free and easy way for customers to try GPUs through a remotely-hosted CPU + GPU server. They can run popular applications, such as AMBER, NAMD or their own code.
HPC High-performance computing is the use of parallel processing for running advanced application programs efficiently, reliably and fast.
Integrated Graphics Graphics hardware that is built into the CPU which shares both CPU hardware resources along with system memory. Only sufficient for very basic applications like email or word processing.
ISV Independent software vendor, typically used to refer to professional software vendors such as Autodesk, Adobe, or Dassault Systèmes.
GloSSArY of TerMS
Scientists studying the physics of the Big Bang need massive computing power. GPUs deliver 125x the performance of other systems.
Scientists using GPU-powered systems were able to crack the genetic code behind a deadly strain of E. coli.
GLOSSARY
TerM DeSCrIPTIoNISV Certifications NVIDIA and ISVs work together to test,validate and guarantee the functionality of the hardware
and software.
Kepler The world’s fastest and most efficient GPU from NVIDIA. NVIDIA products built on the NVIDIA®
Kepler™ architecture have names such as NVIDIA Quadro® K5000 and NVIDIA Tesla® K20 (K for Kepler).
Kepler K10, K20, & K20X Latest generation of NVIDIA Tesla GPU Accelerators and the world’s fastest and most efficient high-performance computing architecture.
libraries Pre-written code and subroutines used by developers when developing software. GPU-accelerated libraries provide “drop-in” acceleration to an application.
Mobile workstation A notebook that includes a Quadro Mobile graphics card for professionals.
Mosaic NVIDIA software designed to manage and optimize multiple displays to create large desktop surfaces or a matrix/mosaic of displays for large-scale display installations.
NVIDIA Maximus Second Generation NVIDIA Maximus™ technology combines the visualization and interactive design capability of Quadro GPUs and the high-performance computing power of NVIDIA Tesla GPUs in a single workstation. Technology that gives workstation users the ability to simultaneously perform complex analysis and visualization on a single machine, using Quadro and Tesla processors.
Nview NVIDIA software designed to manage and optimize your workspace over single or multiple displays.
NVS Graphics cards designed for multi-monitor installations in commercial/office/financial settings. Featured in both commercial PCs and workstations.
Parallel Computing Parallel computing is a form of computation in which many calculations are carried out simultaneously, operating on the principle that large problems can often be divided into smaller ones, which are then solved concurrently (“in parallel”).
Quadro Series of graphics cards designed for professionals to use to accelerate their work in the CAD and DCC markets, featured in workstations.
SPeC ViewPerf SPEC ViewPerf measures the 3D rendering performance of systems running under OpenGL and is typically used for benchmarking purposes.
Speed up The performance differential obtained when running an application on a CPU-only system vs. a CPU + GPU system.
Tesla GPu Graphics processing unit designed from the ground-up to accelerate scientific and technical computing workloads for high-performance computing.
workstation A workstation is a highly tested and qualified PC with enormous capability for hardware upgradeability. It is designed to be highly reliable and serviceable, and can be in the form of a desktop or notebook.
GloSSArY of TerMS CoNTINueD
Tesla GPUs power the Square Kilometer Array, a radio frequency telescope 10,000 times more powerful than any current telescope.
With Maximus technology, designers and engineers can iterate and visualize extremely quickly and accurately, producing more design variations in less time.
Visit www.gpugenius.com to learn more about NVIDIA’s GPU solutions:> Online sales and product training modules> Easy-to-share, customer-ready presentations, solution sheets and videos