Virtual GPU Packaging And Licensing Guide - Nvidia

2y ago
4 Views
1 Downloads
319.41 KB
22 Pages
Last View : 16d ago
Last Download : 2m ago
Upload by : Harley Spears
Transcription

NVIDIA Virtual GPU Software Packaging,Pricing, and Licensing GuideApplication NoteDA-09924-001 v10 February 2022

Document HistoryDA-09924-001 v09VersionDateAuthorsDescription of Change01April 22, 2020CH, SMInitial Release02May 12, 2020CH, SMUpdates to “Perpetual Concurrent User License” and“NVIDIA Education Pricing Program” sections03June 25, 2020CH, RR, MRUpdate NVIDIA Virtual Compute Server subscriptionprice04July 20, 2020NNUpdate NVIDIA Virtual Compute Server abbreviatedname05September 30, 2020MR, RRAdd Ampere GPU support06February 17, 2021MR, VK,NS, SMUpdated new Ampere GPU support, vGPU rebrandingchanges07April 14, 2021NS, SMAdded A10 and A1608June 25, 2021NN, SMUpdated CCU description in Table 1Updated maximum concurrent VMs09August 3, 2021NN, SMAdded subscription pricing to Table 9Updated annual license subscription in Table 10Updated education pricing in Table 1210Feb 1, 2022MRAdded A2, A100X and A30X supportNVIDIA Virtual GPU Software Packaging, Pricing, and Licensing GuideDA-09924-001 v10 2

Table of ContentsTerminology1Overview2General Purchasing InformationVirtual GPU Software Product DetailsVirtual GPU Software ProductsNVIDIA vAppsNVIDIA vPCNVIDIA RTX vWSNVIDIA Virtual Compute ServervGPU Software Products and Entitlement57788888vGPU Software Licensing and Pricing11Enterprise vGPU Software PricingSubscription Concurrent User LicenseSubscription per GPU User LicensePerpetual Concurrent User LicenseNVIDIA Education Pricing Program1111121213Deciding the Right License Based on Capability and Entitlement14NVIDIA vGPU Software License Server17NVIDIA Virtual GPU Software Packaging, Pricing, and Licensing GuideDA-09924-001 v10 3

List of TablesTable 1.Table 2.Table 3.123Table 4.Table 5.33Table 6.Table 7.Table 8.Table 9.Table 10.Table 11.Table 12.Table 13.5691212121316NVIDIA Virtual GPU Software Packaging, Pricing, and Licensing GuideDA-09924-001 v10 4

TerminologyThe following table lists the terms that will be used throughout this guide.Table 1.TerminologyTermMeaningSUMSSupport, Upgrade and Maintenance programPerpetual LicenseA non-expiring, permanent software license that can be used on aperpetual basis without the need to renew. SUMS is required and isavailable in five-year increments. One-year SUMS available only forrenewals.Annual SubscriptionA software license that is active for a fixed period as defined by theterms of the subscription. An annual subscription includes SUMS forthe duration of the license term.License ServerAn application that manages license allocation, installed on a physicalor virtual server.Concurrent User (CCU) LicensingA method of allocating licenses based on the number of VMs that areconcurrently being used. A CCU license allows only one concurrent VMto be hosted per license. For application streaming or session-basedsolutions, each concurrent user requires a CCU license.GPUGraphics processing unitPer GPU LicensingA method of allocating licenses based on the number of GPUs that arebeing used. A per GPU license allows a certain number of concurrentVMs to be hosted per license.NVIDIA Virtual GPU Software Packaging, Pricing, and Licensing GuideDA-09924-001 v10 1

OverviewNVIDIA virtual GPU (vGPU) solutions bring the power of NVIDIA GPUs to virtual desktops, applications,and workstations, accelerating graphics and compute to make virtualized workspaces accessible tocreative and technical professionals working from home offices or anywhere.This guide covers the entitlement, packaging, and licensing of the NVIDIA virtual GPU (vGPU) softwarefamily of products. It is intended to be a quick reference to understand the product portfolio at a highlevel, with the corresponding SKU information. It does not contain detailed product information,which can be obtained from the NVIDIA vGPU website at http://www.nvidia.com/virtualgpu. Thisguide is not intended to replace or contradict the End User License Agreement (EULA). Refer to theEULA for more detailed information.Table 2.NVIDIA Virtual GPU Software Licensed ProductsProductDescriptionNVIDIA Virtual Applications (vApps)For organizations deploying Citrix Virtual Apps and Desktops,RDSH or other app streaming or session-based solutions.Designed for PC level applications and server-based desktops.NVIDIA Virtual PC (vPC)For users who want a virtual desktop but need great userexperience leveraging PC Windows applications, browsers, andhigh-definition video.NVIDIA RTX Virtual Workstation (vWS)For users who want to be able to use remote professionalgraphics applications with full performance on any device,anywhere.NVIDIA Virtual Compute Server (vCS)For compute-intensive server workloads, such as artificialintelligence (AI), deep learning, or high-performance computing(HPC) running on Red Hat Enterprise Linux, Red Hat Virtualization,or other supported KVM-based hypervisors.Note: For compute-intensive server workloads, such as AI, deep learning, or HPC running VMware vSphere,refer to the NVIDIA AI Enterprise Packaging, Pricing and Licensing Guide.NVIDIA vGPU software brings graphics and virtualization capabilities to NVIDIA data centerdeployments and is currently supported on the NVIDIA graphics cards listed in Table 3 through Table6. Find certified servers that are supported by NVIDIA GPUs and vGPU software es/vgpu-certified-servers/.NVIDIA Virtual GPU Software Packaging, Pricing, and Licensing GuideDA-09924-001 v10 2

Virtual GPU Software Product DetailsTable 3.Supported NVIDIA Maxwell Graphics CardsNVIDIA M10NVIDIA M60NVIDIA M6Recommended Use caseNVIDIA vPC – User edNumber of GPUs4 NVIDIA Maxwell GPUs2 NVIDIA Maxwell GPUs1 NVIDIA Maxwell GPUTotal NVIDIA CUDA Cores2,560 (640 per GPU)4,096 (2,048 per GPU)1,536Total Memory Size32 GB GDDR532 GB GDDR58 GB GDDR5(8 GB per GPU)16 GB GDDR516 GB GDDR5(8 GB per GPU)8 GB GDDR58 GB GDDR5Max Power225 W300 WTable 4.100 WSupported NVIDIA Pascal Graphics CardsRecommendedUse caseNVIDIA P4NVIDIA P6NVIDIA P40NVIDIA P100NVIDIA vWSNVIDIA vWS orNVIDIA vPC -NVIDIA vWS -NVIDIA vWSPerformanceOptimized(Mid to high)(Entry to mid)–Blade-Optimized(Mid to high)Number of GPUs1 NVIDIA PascalGP1041 NVIDIA PascalGP1041 NVIDIA PascalGP1021 NVIDIA Pascal GP100Total CUDA Cores2,5602,0483,8403,584Total MemorySize8 GB GDDR516 GB GDDR524 GB GDDR512 GB HBM2/16 GBHBM2Max Power75 W90 W250 W250 W/300 WForm FactorPCIe 3.0 Single SlotMXMPCIe 3.0 Dual SlotSXM2/PCIe 3.0Board Dimensions2.7" 6.6"3.2" 4.1"10.5" 4.4"10.5" 4.4"Cooling SolutionPassiveBare BoardPassivePassiveTable 5.Supported NVIDIA T4, NVIDIA RTX, and NVIDIA V100 Graphics CardsRecommended UsecaseNVIDIA T4NVIDIA RTX 6000NVIDIA RTX 8000NVIDIA V100S/V100SXM2NVIDIA vWS PerformanceNVIDIA vWS -NVIDIA vWS -NVIDIA vWS –Performance OptimizedNVIDIA Virtual GPU Software Packaging, Pricing, and Licensing GuideDA-09924-001 v10 3

Virtual GPU Software Product DetailsOptimized (Entry);vPC - DensityOptimized; vCS –Compute Optimized(High end); vCS –Compute OptimizedNumber of GPUs1 NVIDIA Turing TU1041 NVIDIA Turing TU1021 NVIDIA TuringTU1021 NVIDIA Volta GPUTotal CUDA Cores2,5604,6084,6085,120Tensor Cores320576576640NVIDIA TensorRTCores407272--Total Memory Size16 GB GDDR624 GB GDDR648 GB GDDR632 GB HBM2Max GPU Power70 W260 W / 250 W260 W / 250 W250 W / 300 WForm FactorPCIe 3.0 Single SlotPCIe 3.0 Dual SlotPCIe 3.0 Dual SlotPCIe, SXM2, PerformanceOptimized(High end); vCS –Compute OptimizedPerformanceOptimized; vCS –Compute OptimizedFull Height/LengthBoard Dimensions2.7" 6.6"10.5" 4.4"10.5" 4.4"10.5" 4.4"5.5" 3.1" .5"Cooling SolutionTable 6.PassiveActive or PassiveActive or PassivePassiveSupported NVIDIA Ampere Graphics Cards for compute workloadsNVIDIA HGXA100NVIDIA A100RecommendedUse caseComputeOptimizedComputeOptimizedNumber of GPUs4 NVIDIA A100/ 8NVIDIA A1001 NVIDIA A100FP32 Cores / GPU6,9126,912Tensor Cores /GPU432432TensorRT Cores-Total MemorySize / GPUNVIDIA A30NVIDIA A100XNVIDIA A30XComputeOptimizedData intensive AIand computeData intensive AIand compute1 NVIDIA A301 NVIDIA A100358469123584224432224----40 GBHBM2/80GBHBM2e40 GB HBM2/80GB HBM2e24 GB80 GB24 GBHBM2HBM2eHBM2e77474Max GPU Power /GPU400 W250 W/300 W165W300W230WForm Factor4x SXM4GPUs/8x SXM4PCIe 4.0 DualSlot FHFLPCIe 4.0 2-SlotFHFLPCIe 4.0 2-Slot FHFLPCIe 4.0 2-SlotFHFLMIGInstances/GPUNVIDIA Virtual GPU Software Packaging, Pricing, and Licensing Guide1 NVIDIAA30DA-09924-001 v10 4

Virtual GPU Software Product DetailsBoardDimensionsCooling SolutionTable 7.10.5” 4.4”Passive10.5” 4.4”10.5” 4.4”10.5” 4.4”PassivePassivePassivePassiveSupported NVIDIA Ampere Graphics Cards for mixed workloadsNVIDIA A40NVIDIA A10NVIDIA A16NVIDIA vWS PerformanceOptimized (High end);NVIDIA vWS –Performance Optimized(midrange)VDI: vPC & vWS(entry)Compute OptimizedCompute OptimizedNumber of GPUs1 NVIDIA A401 NVIDIA A104 NVIDIA A161 NVIDIA A2FP32 Cores / GPU10,5729,2164x 12801280Tensor Cores /GPU3362884x 4040TensorRT Cores84724x 1010Total MemorySize / GPU48 GB GDDR624 GB GDDR664GB (4x 16GB)GDDR616 GB GDDR6Max GPU Power /GPU300 W150 W250W60WForm FactorPCIe 4.0 Dual-SlotFHFLPCIe 4.0 Single SlotFHFLPCIe 4.0PCIe 4.01-Slot FHFL1-Slot HHHLBoardDimensions10.5” 4.4”10.5” x 4.4”10.5” x 4.4”2.7” x 6.6”Cooling SolutionPassivePassivePassivePassiveRecommendedUse caseNVIDIA A2MixedProfessional Graphics & Compute(entry)Note: NVIDIA RTX A6000 and NVIDIA RTX A5000 are also supported with NVIDIA virtual GPU software.General Purchasing InformationNVIDIA vGPU software products can be purchased through NVIDIA Preferred Partners and selectserver OEMs. A list of these Preferred Partners and OEMs can be obtained from:http://www.nvidia.com/buygrid.NVIDIA vGPU software products can be purchased as either a perpetual license with a SupportUpdates and Maintenance Subscription (SUMS), or as an annual subscription. The perpetual licensegives the user the right to use the software indefinitely, with no expiration. All NVIDIA vGPU softwareNVIDIA Virtual GPU Software Packaging, Pricing, and Licensing GuideDA-09924-001 v10 5

Virtual GPU Software Product Detailsproducts with perpetual licenses must be purchased in conjunction with a five-year SUMS. A one-yearSUMS is available only for renewals.The annual subscription offering is a more affordable option to allow IT departments to bettermanage the flexibility of license volumes. NVIDIA vGPU software products with an annual subscriptionare bundled with SUMS for the duration of the software’s subscription license.Table 7.General Purchasing InformationEntitlementNVIDIA vGPU Production SUMSMaintenanceAccess to all maintenance releases, defect resolutions, andsecurity patches for flexibility in upgrading as per the NVIDIAVirtual GPU Software Lifecycle PolicyUpgradesAccess to all new major version releases including featureenhancements and new hardware supportLong-term branch maintenanceAvailable for up to 3 years from general availability as per theNVIDIA Virtual GPU Software Lifecycle PolicyDirect supportDirect access to NVIDIA support engineering for timely resolutionof customer-specific issuesSupport availabilityCustomer support available during standard business hours Casesaccepted 24 7Knowledgebase accessWeb supportE-mail supportPhone supportNVIDIA Virtual GPU Software Packaging, Pricing, and Licensing Guide DA-09924-001 v10 6

Virtual GPU Software Product DetailsNVIDIA vGPU software is the industry's most advanced technology for sharing true virtual GPUhardware acceleration between multiple users—without compromising the graphics experience. Thisvirtualization technology ensures complete application compatibility, which means features andexperience are the same as they would be on a physical device.Virtual GPU Software ProductsNVIDIA vGPU desktop and application virtualization solutions are designed to bring the power ofvirtualization to the users who need to be their most productive. vGPU technology ensuresapplication compatibility, meaning any application that can run in a physical desktop environment canrun in a virtual environment. Organizations can now expand their virtualization footprint withoutcompromise.The following NVIDIA vGPU software products are available: NVIDIA Virtual Applications (vApps) NVIDIA Virtual PC (vPC) NVIDIA RTX Virtual Workstation (vWS) NVIDIA Virtual Compute Server (vCS)NVIDIA Virtual GPU Software Packaging, Pricing, and Licensing GuideDA-09924-001 v10 7

Virtual GPU Software Product DetailsNVIDIA vAppsThis product is for organizations deploying Citrix Virtual Apps and Desktops, RDSH or other appstreaming or session-based solutions. Designed to deliver PC Windows applications at fullperformance, NVIDIA vApps allows users to access any Windows application at full performance onany device, anywhere.Windows Server hosted RDSH desktops are also supported by NVIDIA vApps.NVIDIA vPCThis product is ideal for users who want a virtual desktop but need great user experience leveragingPC Windows applications, browsers, and high-definition video. NVIDIA vPC delivers a nativeexperience to users in a virtual environment, allowing them to run all their PC applications at fullperformance.NVIDIA RTX vWSThis product is ideal for mainstream and high-end designers who use powerful 3D content creationapplications such as Dassault CATIA, SOLIDWORKS, 3DExcite, Siemens NX, PTC Creo, SchlumbergerPetrel, or Autodesk Maya. NVIDIA vWS allows users to access their professional graphics applicationswith full features and performance, anywhere, on any device.NVIDIA Virtual Compute ServerThis product is for organizations running compute-intensive server workloads such as ArtificialIntelligence (AI), Deep Learning (DL), and High-Performance Computing (HPC). NVIDIA VirtualCompute Server (vCS) is software that enables the NVIDIA GPU to be virtualized to acceleratecompute-intensive server workloads with features such as error-correcting code (ECC) memory, pageretirement, peer-to-peer CUDA transfers over NVIDIA NVLink , and multiple vGPUs assigned to asingle VM.vGPU Software Products and EntitlementEach NVIDIA vGPU software product includes the feature entitlements listed in Table 8.NVIDIA Virtual GPU Software Packaging, Pricing, and Licensing GuideDA-09924-001 v10 8

Virtual GPU Software Product DetailsTable 8.Feature EntitlementsFeatureNVIDIA vAppsNVIDIA vPCNVIDIA vWSNVIDIA vCSLicense Entitlement Concurrent User (CCU)Per GPU Capability Entitlement Desktop Virtualization RDSH App HostingCompute Virtualization Maximum Displays134Maximum Resolution41280 1024Linux Guest OS Support 2 Windows Guest OS Support RDSH Desktop Hosting 2 5120 2880 (5K) NVLinkGPU Pass Through Support87 4 GB5 GB6 GB7680 4320 (8K)4096 2160 (4K)1 In-situ Graphicsonly 5 6 vGPU Profile Sizes Supported512 MB3 GB Multi-vGPU2 GB ECC and Page Retirement1 GB 4CUDA and OpenCL SupportBare Metal Support NVIDIA RTX Enterprise SoftwareFeaturesOpenGL, DirectX and Vulkan 1 NVIDIA Virtual GPU Software Packaging, Pricing, and Licensing Guide9 DA-09924-001 v10 9

Virtual GPU Software Product Details8 GB10 GB12 GB16 GB20 GB24 GB32 GB40 GB 48 GB80 GBNotes: 1Maximum 10 concurrent VMs per GPU2With packaged NVIDIA vApps license3Applies only to the console display in remote application environments. For details, see Supported GPUs4Review the Virtual GPU Software User Guide for the supported display configurations for each profile5Supported on 8GB 1:1 profile on NVIDIA Maxwell and all profiles on Pascal6ECC support begins with Pascal7Only supported on 1:1 profiles8Only NVIDIA M6 hardware supported as primary display device9Review the Virtual GPU Software User Guide for the vGPU profiles supported on your GPUNVIDIA Virtual GPU Software Packaging, Pricing, and Licensing GuideDA-09924-001 v10 10

vGPU Software Licensing and PricingEnterprise vGPU Software PricingNVIDIA vPC, NVIDIA RTX vWS, and NVIDIA vApps are available on a per Concurrent User (CCU) model.A CCU license is required for every user who is accessing or using the software at any given time,whether an active connection to the virtualized desktop or session is maintained.NVIDIA vCS is available on a per GPU model. A GPU license is required for every GPU that will hostvCS-enabled VMs. A single vCS license enables a maximum of 10 concurrent VMs.NVIDIA vGPU software products can be purchased by enterprises as either perpetual licenses withannual Support Updates and Maintenance Subscription (SUMS), or as an annual subscription. AllNVIDIA vGPU software products with perpetual licenses must be purchased in conjunction with a fiveyear SUMS. A one-year SUMS is available only for renewals. For annual licenses, SUMS is bundled intothe annual license cost.Enterprise vGPU software licensing and pricing are described in the following subsections. Find the fullSKU list here. Pricing is suggested pricing only. Contact your authorized NVIDIA partner for finalpricing. If you are looking to run or host a service by using NVIDIA vGPU software, you need to join theNPN CSP partner program.Subscription Concurrent User LicenseAn annual Enterprise subscription is active for a fixed period as defined by the terms of thesubscription license. To be kept active, the license will need to be renewed at the end of thesubscription period. The subscription license includes the software license and production level SUMSfor the duration of the license subscription period.NVIDIA Virtual GPU Software Packaging, Pricing, and Licensing GuideDA-09924-001 v10 11

Virtual GPU Software Product DetailsTable 9.Concurrent User License Annual Subscription PricingLicenseAnnual Subscription PricingNVIDIA Virtual Applications 10 per CCU subscription 50 per CCU subscriptionNVIDIA Virtual PC4 year subscription - 43.75 per CCU5 year subscription - 40 per CCU 250 per CCU subscriptionNVIDIA RTX Virtual Workstation4 year subscription - 206.25 per CCU5 year subscription - 180 per CCUSubscription per GPU User LicenseAn annual enterprise subscription is active for a fixed period as defined by the terms of thesubscription license. To be kept active, the license will need to be renewed at the end of thesubscription period. The subscription license includes the software license and production-level SUMSfor the duration of the license subscription period.Table 10.GPU User License Annual Subscription PricingLicenseAnnual Subscription PricingNVIDIA Virtual Compute Server 450 per GPU subscriptionPerpetual Concurrent User LicenseA perpetual enterprise license allows for use of the licensed software indefinitely. Users that opt tolicense using this model are required to subscribe to SUMS for five years. The SUMS subscription canbe renewed on a yearly basis after the expiration of the initial subscription.Table 11.Enterprise Perpetual Licensing plus SUMS PricingLicenseNVIDIA Virtual ApplicationsNVIDIA Virtual PCNVIDIA RTX Virtual WorkstationAnnual Licensing plus SUMS Pricing 20 per CCU perpetual license 5 SUMS per year 100 per CCU perpetual license 25 SUMS per year 450 per CCU perpetual licenseNVIDIA Virtual GPU Software Packaging, Pricing, and Licensing GuideDA-09924-001 v10 12

Virtual GPU Software Product Details 100 SUMS per yearNVIDIA Education Pricing ProgramThe NVIDIA Education Pricing Program supports the use of visual computing in teaching and researchinstitutions. The program makes it easy to procure and administer NVIDIA solutions, softwarelicensing, and services for qualified educational institutions and helps reduce their total cost. Formore information on eligibility, review the NVIDIA Education Pricing Program documentation.The Annual Subscription Model is available for NVIDIA vWS and vCS. The Perpetual Licensing Model isavailable only for NVIDIA vWS. Users that opt for the Perpetual Licensing Model are required tosubscribe to SUMS for five years. The SUMS subscription can be renewed on a yearly basis after theexpiration of the initial subscription.Table 12.NVIDIA Education PricingLicensePricing ModelPriceNVIDIA vWSPerpetual Licensing plus SUMS 99 per CCU perpetual license 25 SUMS per yearNVIDIA Virtual Compute ServerAnnual Subscription 50 per CCU subscriptionAnnual Subscription 150 per GPU subscriptionNVIDIA Virtual GPU Software Packaging, Pricing, and Licensing GuideDA-09924-001 v10 13

Deciding the Right License Based onCapability and EntitlementThe following flowchart provides a simple decision tree to help decide which license is required basedon the desired entitlement and capability. If you have further questions or are unable to decide basedon the decision tree, contact NVIDIA vGPU Support at IDIA Virtual GPU Software Packaging, Pricing, and Licensing GuideDA-09924-001 v10 14

Virtual GPU Software Product DetailsDo you use ProfessionalGraphics Apps?YesRequires a vWS LicenseNoAre you using adesktop/workstation OS toserve desktops?YesRequires a vPC LicenseNoAre you serving RDSH Appsor Desktops?YesRequires a vAppsLicenseNoIs your deploymentsessions based?YesNoAre you runningAI/DL/Inference workloads(no graphics)?YesRequires a vCS LicenseNoPlease consult NVIDIAvirtual GPU supportNVIDIA Virtual GPU Software Packaging, Pricing, and Licensing GuideDA-09924-001 v10 15

Virtual GPU Software Product DetailsThe following table summarizes some common use cases of different solutions. This is not an allinclusive list of possible solutions. If you have questions, contact NVIDIA vGPU Support.Table 13.Common Use Cases and SolutionsI am using I need this license Citrix Virtual DesktopsNVIDIA vPC – for PC level applicationsNVIDIA vWS – for workstation/professional 3D use casesVMware Horizon (View)NVIDIA vPC – for PC level applicationsNVIDIA vWS – for each session utilizing a workstation/professional 3Duse caseCitrix Virtual Apps and DesktopsNVIDIA vAppsNVIDIA vWS – for each session utilizing a workstation/professional 3Duse caseVMware Horizon RDSHNVIDIA vAppsNVIDIA vWS – for each session utilizing a workstation/professional 3Duse caseOther RDSH / Session-basedNVIDIA vAppsNVIDIA vWS – for each session utilizing a workstation/professional 3Duse caseMicrosoft RemoteFXNVIDIA vPC – for PC level applicationsVMware Horizon vSGANVIDIA vPC – for PC level applicationsMicrosoft Hyper-V (DDA)NVIDIA vWSRed Hat Enterprise Linux with KVMvCS – for AI/DL/Inference workloadsNVIDIA vWS – for workstation/professional 3D use casesVMware vSphere for AI, DL, andinference workloadsNVIDIA AI EnterpriseNVIDIA Virtual GPU Software Packaging, Pricing, and Licensing GuideDA-09924-001 v10 16

NVIDIA vGPU Software License ServerThe NVIDIA vGPU software license server provides monitoring and reporting on license usage forcapacity planning. It is available for download with the vGPU software packages from the NVIDIALicensing Portal. This license server can be installed on either a physical server or, more likely, adedicated virtual machine.To ensure that the end user experience is not impaired by license overages or connection issues, thevGPU software runs with or without a valid license server connection. The license server allows the ITadministrator to track license usage and determine correct sizing for their environments. Incompliance with the EULA, IT administrators may also use any other method that reliably tracks thesoftware usage to ensure they have enough licenses for their CCU or per GPU usage.NVIDIA Virtual GPU Software Packaging, Pricing, and Licensing GuideDA-09924-001 v10 17

NoticeThis document is provided for information purposes only and shall not be regarded as a warranty of a certain functionality, condition, or quality of a product.NVIDIA Corporation (“NVIDIA”) makes no representations or warranties, expressed or implied, as to the accuracy or completeness of the information containedin this document and assumes no responsibility for any errors contained herein. NVIDIA shall have no liability for the consequences or use of such information orfor any infringement of patents or other rights of third parties that may result from its use. This document is not a commitment to develop, release, or deliver anyMaterial (defined below), code, or functionality.NVIDIA reserves the right to make corrections, modifications, enhancements, improvements, and any other changes to this document, at any time without notice.Customer should obtain the latest relevant information 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 agreedin an individual sales agreement signed by authorized representatives of NVIDIA and customer (“Terms of Sale”). NVIDIA hereby expressly objects to applying anycustomer general terms and conditions with regards to the purchase of the NVIDIA product referenced in this document. No contractual obligations are formedeither directly or indirectly by this document.NVIDIA products are not designed, authorized, or warranted to be suitable for use in medical, military, aircraft, space, or life support equipment, nor in applicationswhere failure or malfunction of the NVIDIA product can reasonably be expected to result in personal injury, death, or property or environmental damage. NVIDIAaccepts 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 ownrisk.NVIDIA makes no representation or warranty that products based on this document will be suitable for any specified use. Testing of all parameters of each productis not necessarily performed by NVIDIA. It is customer’s sole responsibility to evaluate and determine the applicability of any information contained in thisdocument, ensure the product is suitable and fit for the application planned by customer, and perform the necessary testing for the application in order to avoida 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 resultin additional or different conditions and/or requirements beyond those contained in this document. NVIDIA accepts no 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 document or (ii) customerproduct designs.No license, either expressed or implied, is granted under any NVIDIA patent right, copyright, or other NVIDIA intellectual property right under this document.Information published by NVIDIA regarding third-party products or services does not constitute a license from NVIDIA to use such products or services or awarranty or endorsement thereof. Use of such information may require a license from a third party under the patents or other intellectual property rights of thethird party, or a license from NVIDIA under the patents or other intellectual property rights of NVIDIA.Reproduction of information in this document is permissible only if approved in advance by NVIDIA in writing, reproduced without alteration and in full compliancewith all applicable export laws and regulations, and accompanied by all associated conditions, limitations, and notices.THIS DOCUMENT AND ALL NVIDIA DESIGN SPECIFICATIONS, REFERENCE BOARDS, FILES, DRAWINGS, DIAGNOSTICS, LISTS, AND OTHER DOCUMENTS (TOGETHERAND SEPARATELY, “MATERIALS”) ARE BEING PROVIDED “AS IS.” NVIDIA MAKES NO WARRANTIES, EXPRESSED, IMPLIED, STATUTORY, OR OTHERWISE WITHRESPECT TO THE MATERIALS, AND EXPRESSLY DISCLAIMS ALL IMPLIED WARRANTIES OF NONINFRINGEMENT, MERCHANTABILITY, AND FITNESS FOR A PARTICULARPURPOSE. TO THE EXTENT NOT PROHIBITED BY LAW, IN NO EVENT WILL NVIDIA BE LIABLE FOR ANY DAMAGES, INCLUDING WITHOUT LIMITATION ANY DIRECT,INDIRECT, SPECIAL, INCIDENTAL, PUNITIVE, OR CONSEQUENTIAL DAMAGES, HOWEVER CAUSED AND REGARDLESS OF THE THEORY OF LIABILITY, ARISING OUT OFANY USE OF THIS DOCUMENT, EVEN IF NVIDIA HAS BEEN ADVISED OF THE POSSIBILITY OF SUCH DAMAGES. Notwithstanding any damages that customer mightincur for any reason whatsoever, NVIDIA’s aggregate and cumulative liability towards customer for the products described herein shall be limited in accordancewith the Terms of Sale for the product.TrademarksNVIDIA, the NVIDIA logo, CUDA, NVIDIA HGX, NVIDIA Maxwell, NVIDIA Turing, NVIDIA RTX, NVIDIA Volta, NVLink, and TensorRT are trademarks and/or registeredtrademarks of NVIDIA Corporation in the U.S. and other countries. Other company and product names may be trademarks of the respective companies with whichthey are associated.OpenCLOpenCL is a trademark of Apple Inc. used under license to the Khronos Group Inc.Copyright 2020, 2021, 2022 NVIDIA Corporation. All rights reserved.NVIDIA Corporation 2788 San Tomas Expressway, Santa Clara, CA 95051http://www.nvidia.com

This guide covers the entitlement, packaging, and licensing of the NVIDIA virtual GPU (vGPU) software family of products. It is intended to be a quick reference to understand the product portfolio at a high level, with the corresponding SKU infor

Related Documents:

NVIDIA vCS Virtual GPU Types NVIDIA vGPU software uses temporal partitioning and has full IOMMU protection for the virtual machines that are configured with vGPUs. Virtual GPU provides access to shared resources and the execution engines of the GPU: Graphics/Compute , Copy Engines. A GPU hardware scheduler is used when VMs share GPU resources.

OpenCV GPU header file Upload image from CPU to GPU memory Allocate a temp output image on the GPU Process images on the GPU Process images on the GPU Download image from GPU to CPU mem OpenCV CUDA example #include opencv2/opencv.hpp #include <

NVIDIA virtual GPU products deliver a GPU Experience to every Virtual Desktop. Server. Hypervisor. Apps and VMs. NVIDIA Graphics Drivers. NVIDIA Virtual GPU. NVIDIA Tesla GPU. NVIDIA virtualization software. CPU Only VDI. With NVIDIA Virtu

transplant a parallel approach from a single-GPU to a multi-GPU system. One major reason is the lacks of both program-ming models and well-established inter-GPU communication for a multi-GPU system. Although major GPU suppliers, such as NVIDIA and AMD, support multi-GPUs by establishing Scalable Link Interface (SLI) and Crossfire, respectively .

GPU Tutorial 1: Introduction to GPU Computing Summary This tutorial introduces the concept of GPU computation. CUDA is employed as a framework for this, but the principles map to any vendor’s hardware. We provide an overview of GPU computation, its origins and development, before presenting both the CUDA hardware and software APIs. New Concepts

limitation, GPU implementers made the pixel processor in the GPU programmable (via small programs called shaders). Over time, to handle increasing shader complexity, the GPU processing elements were redesigned to support more generalized mathematical, logic and flow control operations. Enabling GPU Computing: Introduction to OpenCL

Possibly: OptiX speeds both ray tracing and GPU devel. Not Always: Out-of-Core Support with OptiX 2.5 GPU Ray Tracing Myths 1. The only technique possible on the GPU is “path tracing” 2. You can only use (expensive) Professional GPUs 3. A GPU farm is more expensive than a CPU farm 4. A

Latest developments in GPU acceleration for 3D Full Wave Electromagnetic simulation. Current and future GPU developments at CST; detailed simulation results. Keywords: gpu acceleration; 3d full wave electromagnetic simulation, cst studio suite, mpi-gpu, gpu technology confere