skip to main content

ThinkSystem NVIDIA H800 PCIe Gen5 GPUs

Product Guide (withdrawn product)

Home
Top
Author
Updated
31 Oct 2023
Form Number
LP1814
PDF size
17 pages, 363 KB

Abstract

The ThinkSystem NVIDIA H800 PCIe Gen5 GPU delivers high performance, scalability, and security for every workload. It uses breakthrough innovations in the NVIDIA Hopper™ architecture to deliver industry-leading conversational AI, speeding up large language models by 30X over the previous generation.

This product guide provides essential presales information to understand the NVIDIA H800 GPU and its key features, specifications, and compatibility. This guide is intended for technical specialists, sales specialists, sales engineers, IT architects, and other IT professionals who want to learn more about the NVIDIA H800 GPU and consider its use in IT solutions.

Withdrawn: The GPUs described in this product guide are now withdrawn

Change History

Changes in the October 31, 2023 update:

  • The H800 GPUs are now withdrawn from marketing.

Introduction

The ThinkSystem NVIDIA H800 PCIe Gen5 GPU delivers high performance, scalability, and security for every workload. It uses breakthrough innovations in the NVIDIA Hopper™ architecture to deliver industry-leading conversational AI, speeding up large language models by 30X over the previous generation.

The H800 is offered in China, Hong Kong and Macau, however the H100 is not offered in these markets. The differences between the H100 and H800 are the NVLink interface speed and the FP64 performance.

The NVIDIA H800 GPU features fourth-generation Tensor Cores and the Transformer Engine with FP8 precision, further extending NVIDIA’s market-leading AI leadership with up to 9X faster training and an incredible 30X inference speedup on large language models. For high-performance computing (HPC) applications, The GPUs triple the floating-point operations per second (FLOPS) of FP64 and add dynamic programming (DPX) instructions to deliver up to 7X higher performance.

The following figure shows the NVIDIA H800 GPU in double-wide PCIe adapter form factor.

ThinkSystem NVIDIA H800 PCIe Gen5 GPU
Figure 1. ThinkSystem NVIDIA H800 PCIe Gen5 GPU

Did you know?

The NVIDIA H800 NVL Tensor Core GPU is optimized for Large Language Model (LLM) Inferences, with its high compute density, high memory bandwidth, high energy efficiency, and unique NVLink architecture.

Part number information

The following table shows the part numbers for the NVIDIA H800 GPU.

Withdrawn: The GPUs listed below are now withdrawn from marketing.

Table 1. Ordering information
Part number Feature code Description
Double-wide PCIe adapters
4X67A89326 BXEJ ThinkSystem NVIDIA H800 NVL 94GB PCIe Gen5 Passive GPU
4X67A86451 BUJW ThinkSystem NVIDIA H800 80GB PCIe Gen5 Passive GPU
NVLink bridge (order 3 per pair of GPUs)
4X67A71309 BG3F ThinkSystem NVIDIA Ampere NVLink 2-Slot Bridge

The PCIe option part numbers includes the following:

  • One GPU with full-height (3U) adapter bracket attached
  • Documentation

Features

The ThinkSystem NVIDIA H800 PCIe Gen5 GPU delivers high performance, scalability, and security for every workload. The GPU uses breakthrough innovations in the NVIDIA Hopper™ architecture to deliver industry-leading conversational AI, speeding up large language models (LLMs) by 30X over the previous generation.

The NVIDIA H800 GPU for mainstream servers comes with a five-year software subscription, including enterprise support, to the NVIDIA AI Enterprise software suite, simplifying AI adoption with the highest performance. This ensures organizations have access to the AI frameworks and tools they need to build accelerated AI workflows such as AI chatbots, recommendation engines, vision AI, and more.

The NVIDIA H800 GPU features fourth-generation Tensor Cores and the Transformer Engine with FP8 precision, further extending NVIDIA’s market-leading AI leadership with up to 9X faster training and an incredible 30X inference speedup on large language models. For high-performance computing (HPC) applications, the GPU triples the floating-point operations per second (FLOPS) of FP64 and adds dynamic programming (DPX) instructions to deliver up to 7X higher performance. With second-generation Multi-Instance GPU (MIG), built-in NVIDIA confidential computing, and NVIDIA NVLink Switch System, the NVIDIA H800 GPU securely accelerates all workloads for every data center from enterprise to exascale.

Key features of the NVIDIA H800 GPU:

  • NVIDIA H800 Tensor Core GPU

    Built with 80 billion transistors using a cutting-edge TSMC 4N process custom tailored for NVIDIA’s accelerated compute needs, H800 is the world’s most advanced chip ever built. It features major advances to accelerate AI, HPC, memory bandwidth, interconnect, and communication at data center scale.

  • Transformer Engine

    The Transformer Engine uses software and Hopper Tensor Core technology designed to accelerate training for models built from the world’s most important AI   model building block, the transformer. Hopper Tensor Cores can apply mixed FP8 and FP16 precisions to dramatically accelerate AI calculations  for transformers.

  • NVLink Switch System

    The NVLink Switch System enables the scaling of multi-GPU input/output (IO) across multiple servers. The system delivers up to 9X higher bandwidth than InfiniBand HDR on the NVIDIA Ampere architecture.

  • NVIDIA Confidential Computing

    NVIDIA Confidential Computing is a built-in security feature of Hopper that makes NVIDIA H800 the world’s first accelerator with confidential computing capabilities. Users can protect the confidentiality and integrity of their data and applications in use while accessing the unsurpassed acceleration of H800 GPUs.

  • Second-Generation Multi-Instance GPU (MIG)

    The Hopper architecture’s second-generation MIG supports  multi-tenant, multi-user configurations in virtualized environments, securely partitioning the GPU into isolated, right-size instances to maximize quality of service (QoS) for 7X more secured tenants.

  • DPX Instructions

    Hopper’s DPX instructions accelerate dynamic programming algorithms by 40X compared to CPUs and 7X compared to NVIDIA Ampere architecture GPUs. This leads to dramatically faster times in disease diagnosis, real-time routing optimizations, and graph analytics.

Technical specifications

The following table lists the GPU processing specifications and performance of the NVIDIA H800 GPU.

Table 2. Specifications of the NVIDIA H800 GPU
Feature H800 80GB PCIe adapter H800 NVL 94GB PCIe adapter
GPU Architecture NVIDIA Hopper NVIDIA Hopper
NVIDIA Tensor Cores 528 fourth-generation Tensor Cores per GPU TBD
NVIDIA CUDA Cores (shading units) 18,432 FP32 CUDA Cores per GPU TBD
Peak FP64 performance 0.8 TFLOPS TBD
Peak FP64 Tensor Core performance 0.8 TFLOPS TBD
Peak FP32 performance 51 TFLOPS TBD
Peak Tensor Float 32 (TF32) performance 756 TFLOPS* TBD
Peak FP16 performance 1,513 TFLOPS* TBD
Peak Bfloat16 (BF16) performance 1,513 TFLOPS* TBD
Peak FP8 performance 3,026 TFLOPS* TBD
INT8 Integer Performance 3,026 TOPS* TBD
GPU Memory 80 GB HBM2e 94 GB HBM3
Memory Bandwidth 2 TB/sec 3.9 TB/sec
ECC Yes Yes
Interconnect Bandwidth NVLink: 400 GB/sec
PCIe Gen5: 128 GB/sec
NVLink: 400 GB/sec
PCIe Gen5: 128 GB/sec
System Interface PCIe Gen 5.0, x16 lanes PCIe Gen 5.0, x16 lanes
Form Factor PCIe full height/length, double width PCIe full height/length, double width
Multi-Instance GPU (MIG) Up to 7 GPU instances, 10GB each Up to 7 GPU instances, 12GB each
Max Power Consumption 350 W 400 W
Thermal Solution Passive Passive
Compute APIs CUDA, DirectCompute, OpenCL, OpenACC CUDA, DirectCompute, OpenCL, OpenACC

* With structural sparsity enabled

Server support

The following tables list the ThinkSystem servers that are compatible.

NVLink server support: The NVLink Ampere bridge is supported with additional NVIDIA A-series and H-series GPUs. As a result, there are additional servers listed as supporting the bridge that don't support the H800 GPU.

Table 3. Server support (Part 1 of 4)
Part Number Description AMD V3 2S Intel V3/V4 4S 8S Intel V3 Multi Node V3/V4 GPU Rich
SR635 V3 (7D9H / 7D9G)
SR655 V3 (7D9F / 7D9E)
SR645 V3 (7D9D / 7D9C)
SR665 V3 (7D9B / 7D9A)
ST650 V3 (7D7B / 7D7A)
SR630 V3 (7D72 / 7D73)
SR650 V3 (7D75 / 7D76)
SR630 V4 (7DG8 / 7DG9)
SR850 V3 (7D97 / 7D96)
SR860 V3 (7D94 / 7D93)
SR950 V3 (7DC5 / 7DC4)
SD535 V3 (7DD8 / 7DD1)
SD530 V3 (7DDA / 7DD3)
SD550 V3 (7DD9 / 7DD2)
SD520 V4 (7DFZ / 7DFY)
SR670 V2 (7Z22 / 7Z23)
SR675 V3 (7D9Q / 7D9R)
SR680a V3 (7DHE)
SR685a V3 (7DHC)
SR780a V3 (7DJ5)
4X67A89326 ThinkSystem NVIDIA H800 NVL 94GB PCIe Gen5 Passive GPU N N N N N N N N N N N N N N N N 8 N N N
4X67A86451 ThinkSystem NVIDIA H800 80GB PCIe Gen5 Passive GPU N N N N N N 3 N 2 4 N N N N N N 8 N N N
4X67A71309 ThinkSystem NVIDIA Ampere NVLink 2-Slot Bridge N N N N N N N N N N N N N N N Y Y N N N
Table 4. Server support (Part 2 of 4)
Part Number Description 1S V3 Edge Super Computing 1S Intel V2 2S Intel V2
ST50 V3 (7DF4 / 7DF3)
ST250 V3 (7DCF / 7DCE)
SR250 V3 (7DCM / 7DCL)
SE350 (7Z46 / 7D1X)
SE350 V2 (7DA9)
SE360 V2 (7DAM)
SE450 (7D8T)
SE455 V3 (7DBY)
SC750 V4 (7DDJ)
SD665 V3 (7D9P)
SD665-N V3 (7DAZ)
SD650 V3 (7D7M)
SD650-I V3 (7D7L)
SD650-N V3 (7D7N)
ST50 V2 (7D8K / 7D8J)
ST250 V2 (7D8G / 7D8F)
SR250 V2 (7D7R / 7D7Q)
ST650 V2 (7Z75 / 7Z74)
SR630 V2 (7Z70 / 7Z71)
SR650 V2 (7Z72 / 7Z73)
4X67A89326 ThinkSystem NVIDIA H800 NVL 94GB PCIe Gen5 Passive GPU N N N N N N N N N N N N N N N N N N N N
4X67A86451 ThinkSystem NVIDIA H800 80GB PCIe Gen5 Passive GPU N N N N N N N N N N N N N N N N N N N N
4X67A71309 ThinkSystem NVIDIA Ampere NVLink 2-Slot Bridge N N N N N N N N N N N N N N N N N N N N
Table 5. Server support (Part 3 of 4)
Part Number Description AMD V1 Dense V2 4S V2 8S 4S V1 1S Intel V1
SR635 (7Y98 / 7Y99)
SR655 (7Y00 / 7Z01)
SR655 Client OS
SR645 (7D2Y / 7D2X)
SR665 (7D2W / 7D2V)
SD630 V2 (7D1K)
SD650 V2 (7D1M)
SD650-N V2 (7D1N)
SN550 V2 (7Z69)
SR850 V2 (7D31 / 7D32)
SR860 V2 (7Z59 / 7Z60)
SR950 (7X11 / 7X12)
SR850 (7X18 / 7X19)
SR850P (7D2F / 2D2G)
SR860 (7X69 / 7X70)
ST50 (7Y48 / 7Y50)
ST250 (7Y45 / 7Y46)
SR150 (7Y54)
SR250 (7Y52 / 7Y51)
4X67A89326 ThinkSystem NVIDIA H800 NVL 94GB PCIe Gen5 Passive GPU N N N N N N N N N N N N N N N N N N N
4X67A86451 ThinkSystem NVIDIA H800 80GB PCIe Gen5 Passive GPU N N N N N N N N N N N N N N N N N N N
4X67A71309 ThinkSystem NVIDIA Ampere NVLink 2-Slot Bridge N N N N N N N N N N N N N N N N N N N
Table 6. Server support (Part 4 of 4)
Part Number Description 2S Intel V1 Dense V1
ST550 (7X09 / 7X10)
SR530 (7X07 / 7X08)
SR550 (7X03 / 7X04)
SR570 (7Y02 / 7Y03)
SR590 (7X98 / 7X99)
SR630 (7X01 / 7X02)
SR650 (7X05 / 7X06)
SR670 (7Y36 / 7Y37)
SD530 (7X21)
SD650 (7X58)
SN550 (7X16)
SN850 (7X15)
4X67A89326 ThinkSystem NVIDIA H800 NVL 94GB PCIe Gen5 Passive GPU N N N N N N N N N N N N
4X67A86451 ThinkSystem NVIDIA H800 80GB PCIe Gen5 Passive GPU N N N N N N N N N N N N
4X67A71309 ThinkSystem NVIDIA Ampere NVLink 2-Slot Bridge N N N N N N N N N N N N

Operating system support

The following table lists the supported operating systems.

Tip: These tables are automatically generated based on data from Lenovo ServerProven.

Table 7. Operating system support for ThinkSystem NVIDIA H800 80GB PCIe Gen5 Passive GPU, 4X67A86451
Operating systems
SR650 V3 (4th Gen Xeon)
SR650 V3 (5th Gen Xeon)
SR675 V3
SR850 V3
SR860 V3
Microsoft Windows 10 N Y N N N
Microsoft Windows 11 N Y N N N
Microsoft Windows Server 2019 Y Y Y Y 2 Y 2
Microsoft Windows Server 2022 Y Y Y Y Y
Red Hat Enterprise Linux 8.6 Y N Y Y Y
Red Hat Enterprise Linux 8.7 Y N Y Y Y
Red Hat Enterprise Linux 8.8 Y Y Y Y Y
Red Hat Enterprise Linux 8.9 Y Y N Y Y
Red Hat Enterprise Linux 9.0 Y N Y Y Y
Red Hat Enterprise Linux 9.1 Y N Y Y Y
Red Hat Enterprise Linux 9.2 Y Y Y Y Y
Red Hat Enterprise Linux 9.3 Y Y N Y Y
SUSE Linux Enterprise Server 15 SP4 Y N Y Y Y
SUSE Linux Enterprise Server 15 SP5 Y Y N Y Y
Ubuntu 20.04.5 LTS N N Y Y Y
Ubuntu 22.04 LTS Y Y 1 Y Y Y
VMware vSphere Hypervisor (ESXi) 7.0 U3 Y Y Y Y Y
VMware vSphere Hypervisor (ESXi) 8.0 Y N N N N
VMware vSphere Hypervisor (ESXi) 8.0 U1 Y N Y Y Y
VMware vSphere Hypervisor (ESXi) 8.0 U2 Y Y Y Y Y

1 Ubuntu 22.04.3 LTS/Ubuntu 22.04.4 LTS

2 For limitation, please refer Support Tip TT1591

NVIDIA GPU software

NVIDIA vGPU Software (vApps, vPC, RTX vWS)

Lenovo offers the following virtualization software for NVIDIA GPUs:

  • Virtual Applications (vApps)

    For organizations deploying Citrix XenApp, VMware Horizon RDSH or other RDSH solutions. Designed to deliver PC Windows applications at full performance. NVIDIA Virtual Applications allows users to access any Windows application at full performance on any device, anywhere. This edition is suited for users who would like to virtualize applications using XenApp or other RDSH solutions. Windows Server hosted RDSH desktops are also supported by vApps.

  • Virtual PC (vPC)

    This product is ideal for users who want a virtual desktop but need great user experience leveraging PC Windows® applications, browsers and high-definition video. NVIDIA Virtual PC delivers a native experience to users in a virtual environment, allowing them to run all their PC applications at full performance.

  • NVIDIA RTX Virtual Workstation (RTX vWS)

    NVIDIA RTX vWS is the only virtual workstation that supports NVIDIA RTX technology, bringing advanced features like ray tracing, AI-denoising, and Deep Learning Super Sampling (DLSS) to a virtual environment. Supporting the latest generation of NVIDIA GPUs unlocks the best performance possible, so designers and engineers can create their best work faster. IT can virtualize any application from the data center with an experience that is indistinguishable from a physical workstation — enabling workstation performance from any device.

The following license types are offered:

  • Perpetual license

    A non-expiring, permanent software license that can be used on a perpetual basis without the need to renew. Each Lenovo part number includes a fixed number of years of Support, Upgrade and Maintenance (SUMS).

  • Annual subscription

    A software license that is active for a fixed period as defined by the terms of the subscription license, typically yearly. The subscription includes Support, Upgrade and Maintenance (SUMS) for the duration of the license term.

  • Concurrent User (CCU)

    A method of counting licenses based on active user VMs. If the VM is active and the NVIDIA vGPU software is running, then this counts as one CCU. A vGPU CCU is independent of the connection to the VM.

The following table lists the ordering part numbers and feature codes.

Table 8. NVIDIA vGPU Software
Part number Feature code
7S02CTO1WW
NVIDIA part number Description
NVIDIA vApps
7S020003WW B1MP 711-VAP001+P3CMI00 NVIDIA vApps SUMS ONLY 5Yr, 1 CCU
7S020004WW B1MQ 711-VAP002+P3CMI12 NVIDIA vApps Subscription License 1 Year, 1 CCU
7S020005WW B1MR 711-VAP002+P3CMI36 NVIDIA vApps Subscription License 3 Years, 1 CCU
7S02003DWW S832 711-VAP002+P3CMI48 NVIDIA vApps Subscription License 4 Years, 1 CCU
7S02003EWW S833 711-VAP002+P3CMI60 NVIDIA vApps Subscription License 5 Years, 1 CCU
7S020046WW SDHB 711-VAP001+P3CMI00 NVIDIA vApps Perpetual License, 1 CCU
NVIDIA vPC
7S020009WW B1MV 711-VPC021+P3CMI00 NVIDIA vPC SUMS 5Yr ONLY, 1 CCU
7S02000AWW B1MW 711-VPC022+P3CMI12 NVIDIA vPC Subscription License 1 Year, 1 CCU
7S02000BWW B1MX 711-VPC022+P3CMI36 NVIDIA vPC Subscription License 3 Years, 1 CCU
7S02003FWW S834 711-VPC022+P3CMI48 NVIDIA vPC Subscription License 4 Years, 1 CCU
7S02003GWW S835 711-VPC022+P3CMI60 NVIDIA vPC Subscription License 5 Years, 1 CCU
7S020047WW SDHC 711-VPC021+P3CMI00 NVIDIA vPC Perpetual License, 1 CCU
NVIDIA RTX vWS
7S02000FWW B1N1 711-DWS021+P3CMI00 NVIDIA RTX vWS SUMS ONLY 5Yr, 1 CCU
7S02000GWW B1N2 711-DWS022+P3CMI12 NVIDIA RTX vWS Subsc Lic 1Yr 1 CCU
7S02000HWW B1N3 711-DWS022+P3CMI36 NVIDIA RTX vWS Subscription License 3 Years, 1 CCU
7S02000XWW S6YJ 711-DWS022+P3CMI48 NVIDIA RTX vWS Subscription License 4 Years, 1 CCU
7S02000YWW S6YK 711-DWS022+P3CMI60 NVIDIA RTX vWS Subscription License 5 Years, 1 CCU
7S02000LWW B1N6 711-DWS021+P3EDI00 NVIDIA RTX vWS EDU SUMS ONLY 5Y, 1CCU
7S02000MWW B1N7 711-DWS022+P3EDI12 NVIDIA RTX vWS EDU Subscription License 1 Year, 1 CCU
7S02000NWW B1N8 711-DWS022+P3EDI36 NVIDIA RTX vWS EDU Subscription License 3 Years, 1 CCU
7S02003BWW S830 711-DWS022+P3EDI48 NVIDIA RTX vWS EDU Subscription License 4 Years, 1 CCU
7S02003CWW S831 711-DWS022+P3EDI60 NVIDIA RTX vWS EDU Subscription License 5 Years, 1 CCU
7S020048WW SDHD 711-DWS021+P3CMI00 NVIDIA RTX vWS Perpetual License, 1 CCU
7S020049WW SDHE 711-DWS021+P3EDI00 NVIDIA RTX vWS EDU Perpetual License, 1 CCU

NVIDIA Omniverse Software (OVE)

NVIDIA Omniverse™ Enterprise is an end-to-end collaboration and simulation platform that fundamentally transforms complex design workflows, creating a more harmonious environment for creative teams.

NVIDIA and Lenovo offer a robust, scalable solution for deploying Omniverse Enterprise, accommodating a wide range of professional needs. This document details the critical components, deployment options, and support available, ensuring an efficient and effective Omniverse experience.

Deployment options cater to varying team sizes and workloads. Using Lenovo NVIDIA-Certified Systems™ and Lenovo OVX nodes which are meticulously designed to manage scale and complexity, ensures optimal performance for Omniverse tasks.

Deployment options include:

  • Workstations: NVIDIA-Certified Workstations with RTX 6000 Ada GPUs for desktop environments.
  • Data Center Solutions: Deployment with Lenovo OVX nodes or NVIDIA-Certified Servers equipped with L40, L40S or A40 GPUs for centralized, high-capacity needs.

NVIDIA Omniverse Enterprise includes the following components and features:

  • Platform Components: Kit, Connect, Nucleus, Simulation, RTX Renderer.
  • Foundation Applications: USD Composer, USD Presenter.
  • Omniverse Extensions: Connect Sample & SDK.
  • Integrated Development Environment (IDE)
  • Nucleus Configuration: Workstation, Enterprise Nucleus Server (supports up to 8 editors per scene); Self-Service Public Cloud Hosting using Containers.
  • Omniverse Farm: Supports batch workloads up to 8 GPUs.
  • Enterprise Services: Authentication (SSO/SSL), Navigator Microservice, Large File Transfer, User Accounts SAML/Account Directory.
  • User Interface: Workstation & IT Managed Launcher.
  • Support: NVIDIA Enterprise Support.
  • Deployment Scenarios: Desktop to Data Center: Workstation deployment for building and designing, with options for physical or virtual desktops. For batch tasks, rendering, and SDG workloads that require headless compute, Lenovo OVX nodes are recommended.

The following part numbers are for a subscription license which is active for a fixed period as noted in the description. The license is for a named user which means the license is for named authorized users who may not re-assign or share the license with any other person.

Table 9. NVIDIA Omniverse Software (OVE)
Part number Feature
7S02CTO1WW
NVIDIA part number Description
7S02003ZWW SCX0 721-OV7006+P3CMI12 NVIDIA Omniverse Enterprise Subscription per GPU, 1 Year
7S020042WW SCX3 721-OV7006+P3CMI36 NVIDIA Omniverse Enterprise Subscription per GPU, 3 Years
7S020044WW SD5T 721-OV7006+P3CMI60 NVIDIA Omniverse Enterprise Subscription per GPU, 5 Year
7S020041WW SCX2 721-OV7006+P3INI12 NVIDIA Omniverse Enterprise Subscription per GPU, INC, 1 Year
7S020040WW SCX1 721-OV7006+P3EDI12 NVIDIA Omniverse Enterprise Subscription per GPU, EDU, 1 Year
7S020043WW SCX4 721-OV7006+P3EDI36 NVIDIA Omniverse Enterprise Subscription per GPU, EDU, 3 Years
7S020045WW SD5U 721-OV7006+P3EDI60 NVIDIA Omniverse Enterprise Subscription per GPU EDU, 5 Year

NVIDIA AI Enterprise Software

Lenovo offers the NVIDIA AI Enterprise (NVAIE) cloud-native enterprise software. NVIDIA AI Enterprise is an end-to-end, cloud-native suite of  AI and data analytics software, optimized, certified, and supported by NVIDIA to run on VMware vSphere and bare-metal  with NVIDIA-Certified  Systems™.  It includes key enabling technologies from NVIDIA for rapid deployment, management, and scaling of AI workloads in the modern hybrid cloud.

NVIDIA AI Enterprise is licensed on a per-GPU basis. NVIDIA AI Enterprise products can be purchased as either a perpetual license with support services, or as an annual or multi-year subscription.

  • The perpetual license provides the right to use the NVIDIA AI Enterprise software indefinitely, with no expiration. NVIDIA AI Enterprise with perpetual licenses must be purchased in conjunction with one-year, three-year, or five-year support services. A one-year support service is also available for renewals.
  • The subscription offerings are an affordable option to allow IT departments to better manage the flexibility of license volumes. NVIDIA AI Enterprise software products with subscription includes support services for the duration of the software’s subscription license

The features of NVIDIA AI Enterprise Software are listed in the following table.

Table 10. Features of NVIDIA AI Enterprise Software (NVAIE)
Features Supported in NVIDIA AI Enterprise
Per GPU Licensing Yes
Compute Virtualization Supported
Windows Guest OS Support No support
Linux Guest OS Support Supported
Maximum Displays 1
Maximum Resolution 4096 x 2160 (4K)
OpenGL and Vulkan In-situ Graphics only
CUDA and OpenCL Support Supported
ECC and Page Retirement Supported
MIG GPU Support Supported
Multi-vGPU Supported
NVIDIA GPUDirect Supported
Peer-to-Peer over NVLink Supported
GPU Pass Through Support Supported
Baremetal Support Supported
AI and Data Science applications and Frameworks Supported
Cloud Native ready Supported

Note: Maximum 10 concurrent VMs per product license

The following table lists the ordering part numbers and feature codes.

Table 11. NVIDIA AI Enterprise Software (NVAIE)
Part number Feature code
7S02CTO1WW
NVIDIA part number Description
AI Enterprise Perpetual License  
7S02001BWW S6YY 731-AI7004+P3CMI60 NVIDIA AI Enterprise Perpetual License and Support per GPU Socket, 5 Years
7S02001EWW S6Z1 731-AI7004+P3EDI60 NVIDIA AI Enterprise Perpetual License and Support per GPU Socket, EDU, 5 Years
AI Enterprise Subscription License  
7S02001FWW S6Z2 731-AI7003+P3CMI12 NVIDIA AI Enterprise Subscription License and Support per GPU Socket, 1 Year
7S02001GWW S6Z3 731-AI7003+P3CMI36 NVIDIA AI Enterprise Subscription License and Support per GPU Socket, 3 Years
7S02001HWW S6Z4 731-AI7003+P3CMI60 NVIDIA AI Enterprise Subscription License and Support per GPU Socket, 5 Years
7S02001JWW S6Z5 731-AI7003+P3EDI12 NVIDIA AI Enterprise Subscription License and Support per GPU Socket, EDU, 1 Year
7S02001KWW S6Z6 731-AI7003+P3EDI36 NVIDIA AI Enterprise Subscription License and Support per GPU Socket, EDU, 3 Years
7S02001LWW S6Z7 731-AI7003+P3EDI60 NVIDIA AI Enterprise Subscription License and Support per GPU Socket, EDU, 5 Years

Find more information in the NVIDIA AI Enterprise Sizing Guide.

NVIDIA HPC Compiler Software

Table 12. NVIDIA HPC Compiler
Part number Feature code
7S09CTO6WW
Description
HPC Compiler Support Services
7S090014WW S924 NVIDIA HPC Compiler Support Services, 1 Year
7S090015WW S925 NVIDIA HPC Compiler Support Services, 3 Years
7S09002GWW S9UQ NVIDIA HPC Compiler Support Services, 5 Years
7S090016WW S926 NVIDIA HPC Compiler Support Services, EDU, 1 Year
7S090017WW S927 NVIDIA HPC Compiler Support Services, EDU, 3 Years
7S09002HWW S9UR NVIDIA HPC Compiler Support Services, EDU, 5 Years
7S090018WW S928 NVIDIA HPC Compiler Support Services - Additional Contact, 1 Year
7S09002JWW S9US NVIDIA HPC Compiler Support Services - Additional Contact, 3 Years
7S09002KWW S9UT NVIDIA HPC Compiler Support Services - Additional Contact, 5 Years
7S090019WW S929 NVIDIA HPC Compiler Support Services - Additional Contact, EDU, 1 Year
7S09002LWW S9UU NVIDIA HPC Compiler Support Services - Additional Contact, EDU, 3 Years
7S09002MWW S9UV NVIDIA HPC Compiler Support Services - Additional Contact, EDU, 5 Years
HPC Compiler Premier Support Services
7S09001AWW S92A NVIDIA HPC Compiler Premier Support Services, 1 Year
7S09002NWW S9UW NVIDIA HPC Compiler Premier Support Services, 3 Years
7S09002PWW S9UX NVIDIA HPC Compiler Premier Support Services, 5 Years
7S09001BWW S92B NVIDIA HPC Compiler Premier Support Services, EDU, 1 Year
7S09002QWW S9UY NVIDIA HPC Compiler Premier Support Services, EDU, 3 Years
7S09002RWW S9UZ NVIDIA HPC Compiler Premier Support Services, EDU, 5 Years
7S09001CWW S92C NVIDIA HPC Compiler Premier Support Services - Additional Contact, 1 Year
7S09002SWW S9V0 NVIDIA HPC Compiler Premier Support Services - Additional Contact, 3 Years
7S09002TWW S9V1 NVIDIA HPC Compiler Premier Support Services - Additional Contact, 5 Years
7S09001DWW S92D NVIDIA HPC Compiler Premier Support Services - Additional Contact, EDU, 1 Year
7S09002UWW S9V2 NVIDIA HPC Compiler Premier Support Services - Additional Contact, EDU, 3 Years
7S09002VWW S9V3 NVIDIA HPC Compiler Premier Support Services - Additional Contact, EDU, 5 Years

Auxiliary power cables

The power cables needed for the H800 SXM GPUs are included with the supported servers.

The H800 PCIe GPU option part number does not ship with auxiliary power cables. Cables are server-specific due to length requirements. For CTO orders, auxiliary power cables are derived by the configurator. For field upgrades, cables will need to be ordered separately as listed in the table below.

Table 13. Auxiliary power cables for H800
Auxiliary power cable needed with the SR650 V3, SR655 V3, SR665 V3, SR665, SR650 V2

SBB7A66338400mm 16-pin (2x6+4) cable
Option:
SR650 V3: 4X67A82883, ThinkSystem SR650 V3 GPU Full Length Thermal Option Kit*
SR655 V3: 4X67A86438, ThinkSystem SR655 V3 GPU Enablement Kit*
SR665 V3: 4X67A85856, ThinkSystem SR665 V3 GPU Full Length Thermal Option Kit*
SR665, SR650 V2: 4X97A85028, ThinkSystem 400mm 2x6+4 GPU Power Cable
Feature: BRWK
SBB: SBB7A66338
Base: SC17B33047
FRU: 03KM846

* The option part numbers are for thermal kits and include other components needed to install the GPU. See the SR650 V3 product guide or SR655 V3 product guide or SR665 V3 product guide for details.

Auxiliary power cable needed with the SR675 V3
SBB7A65299235mm 16-pin (2x6+4) cable
Option
: 4X97A84510, ThinkSystem SR675 V3 Supplemental Power Cable for H100 GPU Option
Feature: BSD2
SBB: SBB7A65299
Base: SC17B39301
FRU: 03LE554
Auxiliary power cable needed with the SR850 V3, SR860 V3
SBB7A72759200mm 16-pin (2x6+4) cable
Option: 4X97A88016, ThinkSystem SR850 V3/SR860 V3 H100 GPU Power Cable Option Kit
Feature: BW28
SBB: SBB7A72759
Base: SC17B40604
FRU: 03LF915
Auxiliary power cable needed with the SR670 V2
SBB7A66339215mm 16-pin (2x6+4) cable
Option
: 4X97A85027, ThinkSystem SR670 V2 H100/L40 GPU Option Power Cable
Feature: BRWL
SBB: SBB7A66339
Base: SC17B33046
FRU: 03KM845

Regulatory approvals

The NVIDIA H800 GPU has the following regulatory approvals:

  • RCM
  • BSMI
  • CE
  • FCC
  • ICES
  • KCC
  • cUL, UL
  • VCCI

Operating environment

The NVIDIA H800 GPU has the following operating characteristics:

  • Ambient temperature
    • Operational: 0°C to 50°C (-5°C to 55°C for short term*)
    • Storage: -40°C to 75°C
  • Relative humidity:
    • Operational: 5-85% (5-93% short term*)
    • Storage: 5-95%

* A period not more than 96 hours consecutive, not to exceed 15 days per year.

Warranty

One year limited warranty. When installed in a Lenovo server, the GPU assumes the server’s base warranty and any warranty upgrades.

Related product families

Product families related to this document are the following:

Trademarks

Lenovo and the Lenovo logo are trademarks or registered trademarks of Lenovo in the United States, other countries, or both. A current list of Lenovo trademarks is available on the Web at https://www.lenovo.com/us/en/legal/copytrade/.

The following terms are trademarks of Lenovo in the United States, other countries, or both:
Lenovo®
ServerProven®
ThinkAgile®
ThinkSystem®

The following terms are trademarks of other companies:

AMD is a trademark of Advanced Micro Devices, Inc.

Intel® and Xeon® are trademarks of Intel Corporation or its subsidiaries.

Linux® is the trademark of Linus Torvalds in the U.S. and other countries.

Microsoft®, Windows Server®, and Windows® are trademarks of Microsoft Corporation in the United States, other countries, or both.

Other company, product, or service names may be trademarks or service marks of others.