ThinkSystem and ThinkAgile GPU Summary Reference Information

Lenovo ThinkSystem servers support GPU technology to accelerate different computing workloads, maximize performance for graphic design, virtualization, artificial intelligence and high performance computing applications in Lenovo servers. This document summarizes the features of the GPUs available for supported ThinkSystem servers and ThinkAgile HX and VX appliances.

Figure 1. ThinkSystem Tesla V100S The following table shows GPUs families and the target workloads

Table 1. GPU families and workloads

Form factor NVIDIA AI and Virtualization AMD AI and Virtualization NVIDIA 3D Graphics Dual slot A100 Instinct MI25 A40 Tesla V100S RTX 8000 Tesla V100 Quadro RTX 6000 Tesla M10 (VDI workloads) Quadro P6000 Quadro RTX 5000

Single slot Tesla V100 FHFL Quadro RTX 4000 Tesla T4 Quadro P2200 Quadro P620

ThinkSystem and ThinkAgile GPU Summary 1 ThinkSystem server support The following table summarizes the ThinkSystem server support for the GPUs. The numbers listed in the server columns represent the number of GPUs supported.

Table 2. ThinkSystem server support

Dense/ E 1S 2S Intel AMD 4S Intel Blade ) ) 3 ) 0 ) 3 ) / 6 G ) ) ) ) ) ) ) ) ) ) ) ) ) ) V ) ) 2 2 Z X 8 3 2 8 4 3 9 2 6 9 9 0 X 1 2 ) 6 0 3 7 2 3 1 D / 5 0 0 0 9 0 0 9 1 7 1 0 / 0 4 1 / / D 2 D 1 9 5 Y X X Y X X X 7 Y Z X X 2 / D Y X 7 7 / 3 5 7 7 7 7 7 7 7 3 7 7 7 7 1 7 Y 7 7 / F ) / / / / / / / / / / / / / ) ) ) ) / / / D Z 7 2 Y W 4 1 7 3 2 8 1 5 6 8 0 8 9 1 1 8 6 5 6 / 5 9 7 7 2 2 5 5 0 0 0 9 0 0 3 9 0 1 D 6 1 2 5 1 1 ( ( 4 8 4 0

7 4 Y Y X X Y X X X Y Y Y D D X X X X X X X Z Y X ( 2 2

7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 Y 7 7 ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( V V ( ( ( ( ( ( ( P

7

( 0 0 0 0 0 0 0 0 0 5 5 5 5 0 0 0 0 0 0 0 0 0 0 0 0 0 5 5 3 5 7 9 3 5 7 3 5 4 6 5 5 6 5 5 6 3 5 5 5 5 0 5 5 1 2 5 5 5 5 6 6 6 6 6 6 6 8 8 8 9 8 8 5 6 5 8 3 Description & 5 2 5 E T T R R T R R R R R R R R R R R R R R R R R D D N N

Part number S S S S S S S S S S S S S S S S S S S S S S S S S S S NVIDIA A100, N N N N N N N N N N N 2 4 N 3 N 3‡ N N N N N N N N N N 4X67A13135 NVIDIA A40, N N N N N N N N N N N N 4† N N N N N N N N N N N N N N 4X67A72593† V100S N N N N N N N N N N N 2 4 N 3 N 3 N N N N N 4 N N N N 32GB, 4X67A13124 NVIDIA Tesla V100 32GB, N N N N N N N N N N N 2 4 N 3 N N N N 2 N N N 2 N N N 4X67A12088 NVIDIA Tesla V100 16GB, N N N N N N N N N N N 2 4 N 3 N 3 N N 2 N N N 2 N N N 4C57A09498 NVIDIA Tesla V100 FHHL, N N N N N N N N N N N 3 N N N N N N N N N N N N N N N 4X67A11524 NVIDIA Tesla T4 16GB, 1 N N N N N N N N N 2 5* 8 3 6 3 8 N N N N 2 8 N N N N 4X67A14926 NVIDIA Tesla M10, N N N N N N N N N N N 2 N N N N N N N N N N N 2 N N N 7C57A02891 AMD Instinct N N N N N N N N N N N 2 N N N N N N N N N N N 2 N N N MI25, 7C57A02897 AMD Radeon Instinct MI25 N N N N N N N N N N N N 4 N N N N N N N N N N N N N N (SR670), 4C57A16224 NVIDIA Quadro RTX 8000, N N N N N N N N N N N 2† 4† N 3† N N N N N N N N N N N N 4X67A65441† NVIDIA Quadro RTX 6000, N N N N N N N N N N N 2† 4 N 3† N N N N N N N N N N N N 4X67A13125 NVIDIA Quadro RTX 5000, N N N N N N N N N N N 2 N N N N N N N N N N N N N N N 4X67A17267 NVIDIA Quadro P6000, N N N N N 2 N N N N N 2 N N N N N N N N N N N N N N N 7C57A02895 NVIDIA Quadro RTX 4000, N N N N N 2 N N N N 1 3 N N N N N N N N N N N N N N N 4X67A14934 NVIDIA Quadro P2200, N N 1 N N 2 N N N N 1 N N N N 1 N N N N N N N N N N N 4X67A14935 NVIDIA Quadro P620, N 1 1 N 1 2 N N N N 3 3 N 3 6 3 8 N N N N N N N N N N 4X67A11584

ThinkSystem and ThinkAgile GPU Summary 2 * The SR650 has support for 5x T4 or 5x P4 GPUs in servers with second-generation Intel Xeon Scalable processors only. SR650 systems originally with first-generation processors have support for up to 3x T4 or 2x P4 GPUs. † Only available via Lenovo Scalable Infrastructure (LeSI). Select "AI & HPC – LeSI Solutions" in the DCSC configurator. See the LeSI product guide for details. ‡ Special Bid only

ThinkAgile HX support The following tables summarizes the ThinkAgile HX appliance and certified node support for the GPUs. The numbers listed in the server columns represent the number of GPUs supported.

Table 3. ThinkAgile HX appliance and certified node GPU support

HX Appliances HX Certified Nodes ) ) ) ) ) ) 4 0 4 4 0 0 ) ) ) ) ) ) ) ) ) ) ) ) ) ) ) 8 9 8 8 9 9 U 0 U 3 3 1 4 4 5 9 9 8 0 0 6 X Y X X Y Y 2 2 2 8 8 8 8 8 9 8 8 8 9 9 9 7 7 7 7 7 7 ( ( ( ( ( ( X X D X X X Y D Y Y D Y Y Y Y

7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 R G C R G C ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ------0 0 0 5 0 0 0 0 0 0 1 1 1 1 6 1 1 1 1 1 1 2 2 2 7 2 2 2 2 2 2 2 2 2 2 7 2 2 2 2 2 2 3 5 3 3 5 7 5 5 5 8 0 3 5 3 3 5 7 5 5 5 8

Part 1 1 3 3 3 3 5 5 7 7 1 1 1 3 3 3 3 5 5 7 7 X X X X X X X X X X X X X X X X X X X X X

number Description H H H H H H H H H H H H H H H H H H H H H 4X67A13135 NVIDIA A100 N N N N N N N N N N N N N N N N N N N N N 4X67A72593 NVIDIA A40 N N N N N N N N N N N N N N N N N N N N N 4X67A13124 NVIDIA Tesla V100S 32GB N N N N N N N N N N N N N N N N N N N N N 4X67A12088 NVIDIA Tesla V100 32GB N N N N 2* N N N N N N N N N N 2* N N N N N 4C57A09498 NVIDIA Tesla V100 16GB N N N N 2* N N N N N N N N N N 2* N N N N N 4X67A11524 NVIDIA Tesla V100 FHHL N N N N 3* N N N N N N N N N N 3* N N N N N 4X67A14926 NVIDIA Tesla T4 N N N 2 5* N N N N N 1 N N N 2 5* N N N N N 7C57A02891 NVIDIA Tesla M10 N N N N 2 N N N N N N N N N N 2 N N N N N 7C57A02897 AMD Radeon Instinct MI25 N N N N N N N N N N N N N N N N N N N N N 4C57A16224 AMD Radeon Instinct MI25 N N N N N N N N N N N N N N N N N N N N N (SR670) 4X67A65441 NVIDIA Quadro RTX 8000 N N N N N N N N N N N N N N N N N N N N N 4X67A13125 NVIDIA Quadro RTX 6000 N N N N N N N N N N N N N N N N N N N N N 4X67A17267 NVIDIA Quadro RTX 5000 N N N N N N N N N N N N N N N N N N N N N 7C57A02895 NVIDIA Quadro P6000 N N N N N N N N N N N N N N N N N N N N N 4X67A14934 NVIDIA Quadro RTX 4000 N N N N N N N N N N N N N N N N N N N N N 4X67A14935 NVIDIA Quadro P2200 N N N N N N N N N N N N N N N N N N N N N 4X67A11584 NVIDIA Quadro P620 N N N N N N N N N N N N N N N N N N N N N * These GPUs are only supported in HX appliances and certified nodes with second-generation Intel Xeon Scalable processors.

ThinkAgile VX support

ThinkSystem and ThinkAgile GPU Summary 3 ThinkAgile VX support The following tables summarizes the ThinkAgile VX appliance and certified node support for the GPUs. The numbers listed in the server columns represent the number of GPUs supported.

Table 4. ThinkAgile VX appliance and certified node GPU support

Certified VX Appliances Nodes ) 2 9 ) ) Y 3 4 7 9 9 (

Y Y e 7 7 ( ( d

o e e ) ) 4 d d 3 N

) ) ) ) ) 9 9 . o o t 3 3 2 4 4 Y Y r N N

9 9 9 9 9 7 7 e . . ( ( t t Y Y Y Y Y

r r C 7 7 7 7 7

G N e e ( ( ( ( ( - - N C C 0 0 0 0 0 0 0 4 2 2 2 2 2 2 2 U U U 3 3 5 7 5 3 5 1 2 2

Part 2 3 3 3 5 7 7 X X X X X X X X X X

number Description V V V V V V V V V V 4X67A13135 NVIDIA A100 N N N N N N N N N N 4X67A72593 NVIDIA A40 N N N N N N N N N N 4X67A13124 NVIDIA Tesla V100S 32GB N N N N N N N N N N 4X67A12088 NVIDIA Tesla V100 32GB N N N N N N N N 2 N 4C57A09498 NVIDIA Tesla V100 16GB N N N N N N N N 2 N 4X67A11524 NVIDIA Tesla V100 FHHL N N 3 N N N N N 3 N 4X67A14926 NVIDIA Tesla T4 16GB N N 5* N N N N N 5* N 7C57A02891 NVIDIA Tesla M10 N N 2 N N N N N 2 N 7C57A02897 AMD Radeon Instinct MI25 N N 2 N N N N N 2 N 4C57A16224 AMD Radeon Instinct MI25 (SR670) N N N N N N N N N N 4X67A65441 NVIDIA Quadro RTX 8000 N N N N N N N N N N 4X67A13125 NVIDIA Quadro RTX 6000 N N N N N N N N N N 4X67A17267 NVIDIA Quadro RTX 5000 N N N N N N N N N N 7C57A02895 NVIDIA Quadro P6000 N N N N N N N N N N 4X67A14934 NVIDIA Quadro RTX 4000 N N N N N N N N N N 4X67A14935 NVIDIA Quadro P2200 N N N N N N N N N N 4X67A11584 NVIDIA Quadro P620 N N 3 N N N N N 3 N * The VX3520-G and VX 2U Certified Node have support for five T4 GPUs in servers with second- generation Intel Xeon Scalable processors only. Servers originally with first-generation processors have support for up to three T4 GPUs.

ThinkSystem and ThinkAgile GPU Summary 4 NVIDIA software Lenovo offers the following software for NVIDIA GPUs: GRID 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 GRID 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. GRID 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 GRID Virtual PC delivers a native experience to users in a virtual environment, allowing them to run all of their PC applications at full performance. Quadro Virtual Datacenter (vDWS): This edition is ideal for mainstream and high-end designers who use powerful 3D content creation applications like Dassault CATIA, SOLIDWORKS, and 3DExcite, Siemens NX, PTC Creo, Schlumberger Petrel, or Autodesk Maya. NVIDIA Quadro Virtual Datacenter Workstation allows users to access their professional graphics applications with full features and performance, anywhere, on 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 GRID software is running, then this counts as one CCU. A GRID CCU is independent of the connection to the VM. The following table lists the ordering part numbers and feature codes.

ThinkSystem and ThinkAgile GPU Summary 5 Table 5. NVIDIA software ordering information

Part number Feature code* Description 7S02000PWW S0DE NVIDIA GRID vApps Perpetual License and SUMS 4Yr, 1 CCU 7S020003WW B1MP NVIDIA GRID vApps Perpetual License and SUMS 5Yr, 1 CCU 7S020004WW B1MQ NVIDIA GRID vApps Subscription License 1Yr, 1 CCU 7S02000QWW S0DF NVIDIA GRID vPC Perpetual License and SUMS 4Yr, 1 CCU 7S020009WW B1MV NVIDIA GRID vPC Perpetual License and SUMS 5Yr, 1 CCU 7S02000AWW B1MW NVIDIA GRID vPC Subscription License 1Yr, 1 CCU 7S02000RWW S0DG NVIDIA Quadro vDWS Perpetual License and SUMS 4Yr, 1 CCU 7S02000FWW B1N1 NVIDIA Quadro vDWS Perpetual License and SUMS 5Yr, 1 CCU 7S02000GWW B1N2 NVIDIA Quadro vDWS Subscription License 1Yr, 1 CCU 7S02000SWW S0DH NVIDIA Quadro vDWS EDU Perpetual License and SUMS 4Yr, 1 CCU 7S02000LWW B1N6 NVIDIA Quadro vDWS EDU Perpetual License and SUMS 5Yr, 1 CCU 7S02000MWW B1N7 NVIDIA Quadro vDWS EDU Subscription License 1Yr, 1 CCU * For CTO orders, use MTM 7S02CTO1WW

ThinkSystem and ThinkAgile GPU Summary 6 ThinkSystem NVIDIA A100 GPU The NVIDIA A100 Tensor Core GPU delivers acceleration at every scale for AI, data analytics, and HPC to tackle the world’s toughest computing challenges. As the engine of the NVIDIA data center platform, A100 can efficiently scale up to thousands of GPUs or, using new Multi-Instance GPU (MIG) technology, can be partitioned into seven isolated GPU instances to accelerate workloads of all sizes. A100’s third-generation Tensor Core technology now accelerates more levels of precision for diverse workloads, speeding time to insight as well as time to market. Part numbers Technical specifications Server support Operating system support Auxiliary power cables

Figure 2. ThinkSystem NVIDIA A100 40GB PCIe 4.0 Passive GPU

Part numbers

Table 6. Ordering information

Part number Feature code Description 4X67A13135 BEL5 ThinkSystem NVIDIA A100 40GB PCIe 4.0 Passive GPU

ThinkSystem and ThinkAgile GPU Summary 7 Technical specifications The following table lists the specifications of the NVIDIA A100 PCIe GPU.

Table 7. A100 specifications

Feature Specification GPU Architecture NVIDIA Ampere NVIDIA Tensor Cores 432 third-generation Tensor Cores per GPU NVIDIA CUDA Cores 6912 FP32 CUDA Cores per GPU Double-Precision Performance FP64: 9.7 TFLOPS FP64 Tensor Core: 19.5 TFLOPS Single-Precision Performance FP32: 19.5 TFLOPS Tensor Float 32 (TF32): 156 TFLOPS, 312 TFLOPS* Half-Precision Performance 312 TFLOPS, 624 TFLOPS* Bfloat16 312 TFLOPS, 624 TFLOPS* Integer Performance INT8: 624 TOPS, 1,248 TOPS* INT4: 1,248 TOPS, 2,496 TOPS* GPU Memory 40 GB HBM2 Memory Bandwidth 1.6 TB/sec ECC Yes Interconnect Bandwidth 32 GB/sec System Interface PCIe Gen 4, x16 lanes Form Factor PCIe full height/length, double width Multi-Instance GPU (MIG) Up to 7 GPU instances Max Power Consumption 250 W Thermal Solution Passive Compute APIs CUDA, DirectCompute, OpenCL, OpenACC * With structural sparsity enabled

Server support The following table lists the ThinkSystem servers that are compatible. The numbers listed in the server columns represent the number of GPUs supported.

Table 8. ThinkSystem server support

Dense/ E 1S Intel 2S Intel AMD 4S Intel Blade ) ) 3 ) 0 ) 3 / ) 6 G ) ) ) ) ) ) ) ) ) ) ) ) ) ) V ) ) 2 2 Z X 8 3 2 8 4 3 9 2 6 9 9 0 X 1 2 ) 6 0 3 7 2 3 1 D / 5 0 0 0 9 0 0 9 1 7 1 0 / 0 4 1 / / D 2 D 1 9 5 Y X X Y X X X 7 Y Z X X 2 / D Y X 7 / 7 3 5 7 7 7 7 7 7 7 3 7 7 7 7 1 7 Y 7 7 / F ) / / / / / / / / / / / / / ) ) ) ) / / / D Z 7 2 Y W 4 1 7 3 2 8 1 5 6 8 0 8 9 1 1 8 6 5 6 / 5 9 7 7 2 2 5 5 0 0 0 9 0 0 3 9 0 1 D 6 1 2 5 1 1 ( ( 4 8 4 0

7 4 Y Y X X Y X X X Y Y Y D D X X X X X X X Z Y X ( 2 2

7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 Y 7 7 ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( V V ( ( ( ( ( ( ( P

7

( 0 0 0 0 0 0 0 0 0 5 5 5 5 0 0 0 0 0 0 0 0 0 0 0 0 0 5 5 3 5 7 9 3 5 7 3 5 4 6 5 5 6 5 5 6 3 5 5 5 5 0 5 5 1 2 5 5 5 5 6 6 6 6 6 6 6 8 8 8 9 8 8 5 6 5 8 3 5 2 5 E T T R R T R R R R R R R R R R R R R R R R R D D N N

Description & Part number S S S S S S S S S S S S S S S S S S S S S S S S S S S NVIDIA A100, 4X67A13135 N N N N N N N N N N N 2 4 N 3 N 3* N N N N N N N N N N * Supported only via Special Bid. Please contact your Lenovo sales representative for details.

ThinkSystem and ThinkAgile GPU Summary 8 Operating system support The following table lists the supported operating systems.

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

Table 9. Operating system support for ThinkSystem NVIDIA A100 40GB PCIe Gen4 Passive GPU, 4X67A13135 ) ) ) ) 2 2 1 1

n n n n e e e e G G G G ( ( ( (

5 0 0 0 0 5 5 7 5 7 6 6 6 6 6 R R R R R

Operating systems S S S S S Microsoft Windows Server 2016 Y Y N Y N Microsoft Windows Server 2019 Y Y N Y N Red Hat Enterprise Linux 7.7 Y Y Y Y Y Red Hat Enterprise Linux 7.8 Y Y Y Y Y Red Hat Enterprise Linux 8.1 Y Y Y Y Y Red Hat Enterprise Linux 8.2 Y Y Y Y Y SUSE Linux Enterprise Server 12 SP2 N N N Y N VMware vSphere Hypervisor (ESXi) 6.5 N N N Y N VMware vSphere Hypervisor (ESXi) 6.5 U1 N N N Y N VMware vSphere Hypervisor (ESXi) 6.5 U2 N Y N Y N VMware vSphere Hypervisor (ESXi) 6.5 U3 Y Y N Y N VMware vSphere Hypervisor (ESXi) 6.7 N N N Y N VMware vSphere Hypervisor (ESXi) 6.7 U1 N Y N Y N VMware vSphere Hypervisor (ESXi) 6.7 U2 N Y N Y N VMware vSphere Hypervisor (ESXi) 6.7 U3 Y Y Y Y Y VMware vSphere Hypervisor (ESXi) 7.0 Y Y Y Y Y VMware vSphere Hypervisor (ESXi) 7.0 U1 Y Y Y Y Y

Auxiliary power cables The A100 option part number does not ship with auxiliary power cables. For use in the SR670, the needed power cables are ordered with the server as listed in the table below. For the SR650, the needed cable is part of ThinkSystem SR650 GPU Cable Kit, 4XH7A08794.

ThinkSystem and ThinkAgile GPU Summary 9 Table 10. Auxiliary power cables for A100 (click images to show larger versions)

Auxiliary power cables supplied with the SR670 (configure-to-order) 900mm SR670 Cage 1 power cable Feature: B3Y3 SBB: SBB7A10375 Base: SC17A10876 FRU: 01PG448

Notes:

Feature & SBB also include PCIe data cable Cable also supplies power to the riser

1100mm SR670 Cage 2 power cable Feature: B3Y2 SBB: SBB7A10374 Base: SC17A10863 FRU: 01PG426

Notes:

Feature & SBB also include PCIe data cable Cable also supplies power to the riser

Auxiliary power cables supplied with the SR655 (configure-to-order only) 350mm 8pin (2x4) cable Server support: SR655 (Riser 1 or Riser 2) Feature: B5T5 SBB: SBB7A14640 Base: SC17A50848 FRU: 02JK011

250mm 8pin (2x4) cable Server support: SR655 (Riser 3) Feature: B5TS SBB: SBB7A10974 Base: SC17A50844 FRU: 02JK010

Auxiliary power cables needed with the SR650 (configure-to-order or field upgrade) 300mm 8pin (2x4) cable Option: 4XH7A08794, ThinkSystem SR650 GPU Cable Kit Feature: AUSR SBB: SBB7A00299 Base: SC17A02296 FRU: 01KN066

ThinkSystem and ThinkAgile GPU Summary 10 ThinkSystem NVIDIA A40 GPU The NVIDIA A40 is a powerful data center GPU for visual computing, delivering high performance and capabilities to professionals for graphics-based workloads such as ray traced rendering, high-performance virtual , simulation, 3D design, VR, and virtual production. The A40 GPU is a graphics-based virtualization solution for designers, engineers, scientists, and creatives that need this performance from anywhere in the world. Part numbers Technical specifications Server support Auxiliary power cables

Figure 3. ThinkSystem NVIDIA A40 48GB PCIe Gen4 Passive GPU

Part numbers

Table 11. Ordering information

Part number Feature code Description 4X67A72593* BEL4 ThinkSystem NVIDIA A40 48GB PCIe Gen4 Passive GPU * The NVIDIA A40 is available via Lenovo Scalable Infrastructure, LeSI. Select "AI & HPC – LeSI Solutions" in the DCSC configurator. See the LeSI product guide for details.

ThinkSystem and ThinkAgile GPU Summary 11 Technical specifications The following table lists the specifications of the NVIDIA A40 PCIe GPU.

Table 12. A40 specifications

Feature Specification GPU Architecture NVIDIA Ampere NVIDIA CUDA Parallel 10,752 Processing Cores NVIDIA Tensor Cores (3rd gen) 336 NVIDIA RT Cores (2nd Gen) 84 Peak FP32 performance (non- 37.4 TFLOPS Tensor) Peak FP16 Tensor performance 149.7 TFLOPS, 299.4 TFLOPS* with FP16 Accumulate Peak Tensor Float 32 (TF32) 74.8 TFLOPS, 149.6 TFLOPS* performance Peak Bfloat16 (BF16) 149.7 TFLOPS, 299.4 TFLOPS* performance with FP32 Accumulate Peak Integer Performance INT8: 299.3 TOPS, 598.6 TOPS* INT4: 598.7 TOPS, 1197.4 TOPS* RT Core performance 73.1 TFLOPS GPU Memory 48 GB GDDR6 Memory Bandwidth 696 GB/s ECC Yes Interconnect Bandwidth NVIDIA NVLink 112.5 GB/s (bidirectional) PCIe Gen4 x16 31.5 GB/s (bidirectional) System Interface PCIe Gen 4, x16 lanes Form Factor PCIe full height/length, double width (10.5" x 4.4") Multi-Instance GPU (MIG) No support Max Power Consumption 300 W Thermal Solution Passive vGPU Software Support NVIDIA GRID®, NVIDIA Quadro® Virtual Data Center Workstation, NVIDIA Virtual Compute Server vGPU Profiles Supported 1 GB, 2 GB, 3 GB, 4 GB, 6 GB, 8 GB, 12 GB, 16 GB, 24 GB, 48 GB Display connectors 3x DisplayPort 1.4 (disabled by default, not supported in the SR670 server) Graphics APIs DirectX 12.07, Model 5.17, OpenGL 4.68, Vulkan 1.18 Compute APIs CUDA, DirectCompute, OpenCL, OpenACC * With structural sparsity enabled

ThinkSystem and ThinkAgile GPU Summary 12 Server support The following table lists the ThinkSystem servers that are compatible. The numbers listed in the server columns represent the number of GPUs supported.

Table 13. ThinkSystem server support

Dense/ E 1S Intel 2S Intel AMD 4S Intel Blade ) ) 3 ) 0 ) 3 ) / 6 G ) ) ) ) ) ) ) ) ) ) ) ) ) ) V ) ) 2 2 Z X 8 3 2 8 4 3 9 2 6 9 9 0 X 1 2 ) 6 0 3 7 2 3 1 D / 5 0 0 0 9 0 0 9 1 7 1 0 / 0 4 1 / / D 2 D 1 9 5 Y X X Y X X X 7 Y Z X X 2 / D Y X 7 7 / 3 5 7 7 7 7 7 7 7 3 7 7 7 7 1 7 Y 7 7 / F ) / / / / / / / / / / / / / ) ) ) ) / / / D Z 7 2 Y W 4 1 7 3 2 8 1 5 6 8 0 8 9 1 1 8 6 5 6 / 5 9 7 7 2 2 5 5 0 0 0 9 0 0 3 9 0 1 D 6 1 2 5 1 1 ( ( 4 8 4 0

7 4 Y Y X X Y X X X Y Y Y D D X X X X X X X Z Y X ( 2 2

7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 Y 7 7 ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( V V ( ( ( ( ( ( ( P

7

( 0 0 0 0 0 0 0 0 0 5 5 5 5 0 0 0 0 0 0 0 0 0 0 0 0 0 5 5 3 5 7 9 3 5 7 3 5 4 6 5 5 6 5 5 6 3 5 5 5 5 0 5 5 1 2 5 5 5 5 6 6 6 6 6 6 6 8 8 8 9 8 8 5 6 5 8 3 5 2 5 E T T R R T R R R R R R R R R R R R R R R R R D D N N

Description & Part number S S S S S S S S S S S S S S S S S S S S S S S S S S S NVIDIA A40, 4X67A72593† N N N N N N N N N N N N 4†* N N N N N N N N N N N N N N † Only available via Lenovo Scalable Infrastructure (LeSI). Select "AI & HPC – LeSI Solutions" in the DCSC configurator. See the LeSI product guide for details. * DisplayPort ports not supported in the SR670

ThinkSystem and ThinkAgile GPU Summary 13 Auxiliary power cables The A40 option part number does not ship with auxiliary power cables. For use in the SR670, the needed power cables are ordered with the server as listed in the table below.

Table 14. Auxiliary power cables for A40 (click images to show larger versions)

Auxiliary power cables supplied with the SR670 (configure-to-order) 900mm SR670 Cage 1 power cable Feature: B3Y3 SBB: SBB7A10375 Base: SC17A10876 FRU: 01PG448

Notes:

Feature & SBB also include PCIe data cable Cable also supplies power to the riser

1100mm SR670 Cage 2 power cable Feature: B3Y2 SBB: SBB7A10374 Base: SC17A10863 FRU: 01PG426

Notes:

Feature & SBB also include PCIe data cable Cable also supplies power to the riser

ThinkSystem and ThinkAgile GPU Summary 14 ThinkSystem NVIDIA Tesla V100S GPU The NVIDIA V100S Tensor Core GPU is a powerful accelerator for , machine learning, high performance computing (HPC), and graphics. The V100S GPU offers improved performance over the V100, featuring a ~25% increase in memory bandwidth and higher FLOPS. Introduction Part numbers Technical specifications Server support Operating system support Auxiliary power cables

Introduction The NVIDIA Tesla V100S GPU Accelerator for PCIe is a dual-slot 10.5 inch PCI Express Gen3 card with a single NVIDIA Volta GV100 (GPU), 32GB HBM2 memory with native ECC support. It uses a passive heat sink for cooling. The Tesla V100S GPU supports double precision (FP64), single precision (FP32) and half precision (FP16) compute tasks, unified virtual memory and page migration engine.

Figure 4. ThinkSystem NVIDIA Tesla V100S

Part numbers

Table 15. Ordering information

Part number Feature code Description 4X67A13124 BB2E ThinkSystem NVIDIA Tesla V100S 32GB PCIe Passive GPU

ThinkSystem and ThinkAgile GPU Summary 15 Technical specifications The following table lists the specifications of the NVIDIA Tesla V100S GPU.

Table 16. V100S specifications

Feature Specification GPU Architecture NVIDIA Volta NVIDIA Tensor Cores 640 NVIDIA CUDA Cores 5,120 Double-Precision Performance 8.2 TFLOPS Single-Precision Performance 16.4 TFLOPS Tensor Performance 130 TFLOPS GPU Memory 32 GB HBM2 Memory Bandwidth 1,134 GB/sec ECC Yes Interconnect Bandwidth 32 GB/sec System Interface PCIe Gen3 Form Factor PCIe Full Height/Length Max Power Consumption 250 W Thermal Solution Passive Compute APIs CUDA, DirectCompute, OpenCL, OpenACC

Server support The following table summarizes the ThinkSystem server support for the GPUs. The numbers listed in the server columns represent the number of GPUs supported.

Table 17. ThinkSystem server support

Dense/ E 1S Intel 2S Intel AMD 4S Intel Blade ) ) 3 ) 0 ) 3 ) / G 6 ) ) ) ) ) ) ) ) ) ) ) ) ) ) V ) ) 2 2 X Z 8 3 2 8 4 3 9 2 6 9 9 0 X 1 2 6 0 ) 3 2 7 3 1 D / 5 0 0 0 9 0 0 9 1 7 1 0 / 4 1 / / 0 D 2 D 1 9 Y X X Y X X X 7 Y Z X X 2 5 / D Y X 7 7 / 3 5 7 7 7 7 7 7 7 3 7 7 7 7 1 7 7 7 Y / F ) / / / / / / / / / / / / / ) ) ) ) / / / D Z 7 2 Y W 4 1 7 3 2 8 1 5 6 8 0 8 9 1 1 8 6 5 6 / 5 9 7 7 2 2 5 5 0 0 0 9 0 0 3 9 0 1 D 6 1 2 5 1 1 ( ( 4 8 4 0

7 4 Y Y X X Y X X X Y Y Y D D X X X X X X X Z Y X ( 2 2

7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 Y 7 7 ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( V V ( ( ( ( ( ( ( P

7

( 0 0 0 0 0 0 0 0 0 5 5 5 5 0 0 0 0 0 0 0 0 0 0 0 0 0 5 5 3 5 7 9 3 5 7 3 5 4 6 5 5 6 5 5 6 3 5 5 5 5 0 5 5 1 2 5 5 5 5 6 6 6 6 6 6 6 8 8 8 9 8 8 5 6 5 8 3 5 2 5 E T T R R T R R R R R R R R R R R R R R R R R D D N N

Description & Part number S S S S S S S S S S S S S S S S S S S S S S S S S S S NVIDIA Tesla V100S 32GB, N N N N N N N N N N N 2 4 N 3 N 3 N N N N N 4 N N N N 4X67A13124

ThinkSystem and ThinkAgile GPU Summary 16 Operating system support The following table lists the supported operating systems.

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

Table 18. Operating system support for ThinkSystem NVIDIA Tesla V100S 32GB PCIe Passive GPU, 4X67A13124 ) ) ) ) 2 2 1 1

n n n n e e e e 2 G G G G ( ( V ( (

5 5 0 0 0 0 0 5 6 5 7 6 5 7 6 6 6 6 8 6 6 R R R R R R R

Operating systems S S S S S S S Microsoft Windows Server 2012 R2 N N N N N Y N Microsoft Windows Server 2016 Y Y Y N Y Y N Microsoft Windows Server 2019 Y Y Y N Y Y N Red Hat Enterprise Linux 7.5 N N N Y N Y Y Red Hat Enterprise Linux 7.6 Y N Y Y N Y Y Red Hat Enterprise Linux 7.7 Y N Y Y N Y Y Red Hat Enterprise Linux 7.8 Y N Y Y N Y Y Red Hat Enterprise Linux 7.9 N N N N Y N N Red Hat Enterprise Linux 8.0 Y N Y Y N Y Y Red Hat Enterprise Linux 8.1 Y N Y Y N Y Y Red Hat Enterprise Linux 8.2 Y N Y Y Y Y Y SUSE Linux Enterprise Server 12 SP4 Y N Y N N Y N SUSE Linux Enterprise Server 12 SP4 with Xen Y N Y N N Y N SUSE Linux Enterprise Server 12 SP5 N N N N Y N N SUSE Linux Enterprise Server 12 SP5 with Xen N N N N Y N N SUSE Linux Enterprise Server 15 N N Y N N Y N SUSE Linux Enterprise Server 15 SP1 Y Y Y Y N Y Y SUSE Linux Enterprise Server 15 SP1 with Xen Y Y Y Y N Y Y SUSE Linux Enterprise Server 15 SP2 N N N N Y N N SUSE Linux Enterprise Server 15 SP2 with Xen N N N N Y N N SUSE Linux Enterprise Server 15 with Xen N N Y N N Y N VMware vSphere Hypervisor (ESXi) 6.5 N N N N N Y N VMware vSphere Hypervisor (ESXi) 6.5 U1 N N N N N Y N VMware vSphere Hypervisor (ESXi) 6.5 U2 N N Y N N Y N VMware vSphere Hypervisor (ESXi) 6.5 U3 Y N Y N N Y N VMware vSphere Hypervisor (ESXi) 6.7 N N N N N Y N VMware vSphere Hypervisor (ESXi) 6.7 U1 N N Y N N Y N VMware vSphere Hypervisor (ESXi) 6.7 U2 N N Y N N Y N VMware vSphere Hypervisor (ESXi) 6.7 U3 Y Y Y Y N Y Y VMware vSphere Hypervisor (ESXi) 7.0 Y N Y Y N Y Y

Auxiliary power cables

ThinkSystem and ThinkAgile GPU Summary 17 Auxiliary power cables The V100S option part numbers ship with two auxiliary power cables. These cables are needed when the GPU is installed in specific servers, as listed in the following table. The second cable listed, SBB7A16972, is needed when the GPU is installed in the SR670 and is used in conjunction with the Cage 1 and Cage 2 power cables that are included with the SR670. The table also lists the additional cables needed when the GPU is installed in the SR670 and SR665 servers.

Table 19. Auxiliary power cables for V100S (click images to show larger versions)

Auxiliary power cables included with the GPU option part number 300mm 8pin (2x4) cable Server support: SR650† SBB: SBB7A00299 Base: SC17A02296 FRU: 01KN066

100mm PCIe 6pin (2x3) + PCIe 8pin (2x4): Server support: SR670 Option: 4Z57A26300* SBB: SBB7A16972 Base: SC17A10850 FRU: 01PG437

Note: This cable is used in conjunction with either SBB7A10375 or SBB7A10374 listed below.

Auxiliary power cables supplied with the SR655 (configure-to-order only) 350mm 8pin (2x4) cable Server support: SR655 (Riser 1 or Riser 2) Feature: B5T5 SBB: SBB7A14640 Base: SC17A50848 FRU: 02JK011

250mm 8pin (2x4) cable Server support: SR655 (Riser 3) Feature: B5TS SBB: SBB7A10974 Base: SC17A50844 FRU: 02JK010

Auxiliary power cables supplied with the SR670 (configure-to-order)

ThinkSystem and ThinkAgile GPU Summary 18 900mm SR670 Cage 1 power cable Feature: B3Y3 SBB: SBB7A10375 Base: SC17A10876 FRU: 01PG448

Notes:

Feature & SBB also include PCIe data cable Cable also supplies power to the riser

1100mm SR670 Cage 2 power cable Feature: B3Y2 SBB: SBB7A10374 Base: SC17A10863 FRU: 01PG426

Notes:

Feature & SBB also include PCIe data cable Cable also supplies power to the riser

Auxiliary power cables needed with the SR665 (part number or feature code) 360mm 8pin (2x4) cable Option: 4M17A11759, ThinkSystem 2U GPU Thermal Option Kit Feature: BAD8 SBB: SBB7A21691 Base: SC17A59596 FRU: 02YE420

Note: The option part number includes other components needed to install the GPU. See theS R665 product guide for details.

ThinkSystem and ThinkAgile GPU Summary 19 * This cable is also available as a separate option part number for customers that buy the V100S directly from NVIDIA and wish to use the GPU in the SR670 server. † For customers who buy the GPU directly from NVIDIA, this cable is also included in the ThinkSystem SR650 GPU Cable Kit, 4XH7A08794

ThinkSystem and ThinkAgile GPU Summary 20 ThinkSystem NVIDIA Tesla V100 GPU The NVIDIA V100 Tensor Core GPU is the world’s most powerful accelerator for deep learning, machine learning, high performance computing (HPC), and graphics. Powered by NVIDIA Volta, a single V100 Tensor Core GPU offers the performance of nearly 32 CPUs—enabling researchers to tackle challenges that were once unsolvable. The V100 won MLPerf, the first industry-wide AI benchmark, validating itself as the world’s most powerful, scalable, and versatile computing platform. Introduction Part numbers Technical specifications Server support Operating system support Auxiliary power cables

Introduction The NVIDIA Tesla V100 GPU Accelerator for PCIe is a dual-slot 10.5 inch PCI Express Gen3 card with a single NVIDIA Volta GV100 graphics processing unit (GPU), 16GB or 32GB HBM2 memory with native ECC support. It uses a passive heat sink for cooling. The Tesla V100 GPU supports double precision (FP64), single precision (FP32) and half precision (FP16) compute tasks, unified virtual memory and page migration engine.

Figure 5. ThinkSystem NVIDIA Tesla V100

Part numbers

Table 20. Ordering information

Part number Feature code Description 4X67A12088 B34S ThinkSystem NVIDIA Tesla V100 32GB

ThinkSystem and ThinkAgile GPU Summary 21 Technical specifications The following table lists the specifications of the NVIDIA Tesla V100 GPU.

Table 21. V100 specifications

Feature Specification GPU Architecture NVIDIA Volta NVIDIA Tensor Cores 640 NVIDIA CUDA Cores 5,120 Double-Precision Performance 7 TFLOPS Single-Precision Performance 14 TFLOPS Tensor Performance 112 TFLOPS GPU Memory 4X67A12088: 32 GB HBM2 Memory Bandwidth 900 GB/sec ECC Yes Interconnect Bandwidth 32 GB/sec System Interface PCIe Gen3 Form Factor PCIe Full Height/Length Max Power Consumption 250 W Thermal Solution Passive Compute APIs CUDA, DirectCompute, OpenCL, OpenACC

Server support The following table lists the ThinkSystem servers that are compatible. The numbers listed in the server columns represent the number of GPUs supported.

Table 22. ThinkSystem server support

Dense/ E 1S Intel 2S Intel AMD 4S Intel Blade ) ) 3 ) 0 ) 3 ) / G 6 ) ) ) ) ) ) ) ) ) ) ) ) ) ) V ) ) 2 2 X Z 8 3 2 8 4 3 9 2 6 9 9 0 X 1 2 6 0 ) 3 2 7 3 1 D / 5 0 0 0 9 0 0 9 1 7 1 0 / 4 1 / / 0 D 2 D 1 9 Y X X Y X X X 7 Y Z X X 2 5 / D Y X 7 7 / 3 5 7 7 7 7 7 7 7 3 7 7 7 7 1 7 7 7 Y / F ) / / / / / / / / / / / / / ) ) ) ) / / / D Z 7 2 Y W 4 1 7 3 2 8 1 5 6 8 0 8 9 1 1 8 6 5 6 / 5 9 7 7 2 2 5 5 0 0 0 9 0 0 3 9 0 1 D 6 1 2 5 1 1 ( ( 4 8 4 0

7 4 Y Y X X Y X X X Y Y Y D D X X X X X X X Z Y X ( 2 2

7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 Y 7 7 ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( V V ( ( ( ( ( ( ( P

7

( 0 0 0 0 0 0 0 0 0 5 5 5 5 0 0 0 0 0 0 0 0 0 0 0 0 0 5 5 3 5 7 9 3 5 7 3 5 4 6 5 5 6 5 5 6 3 5 5 5 5 0 5 5 1 2 5 5 5 5 6 6 6 6 6 6 6 8 8 8 9 8 8 5 6 5 8 3 5 2 5 E T T R R T R R R R R R R R R R R R R R R R R D D N N

Description & Part number S S S S S S S S S S S S S S S S S S S S S S S S S S S NVIDIA Tesla V100 32GB, N N N N N N N N N N N 2 4 N 3 N N N N 2 N N N 2 N N N 4X67A12088

ThinkSystem and ThinkAgile GPU Summary 22 Operating system support The following tables list the supported operating systems.

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

Table 23. Operating system support for ThinkSystem NVIDIA Tesla V100 32GB GPU, PCIe, 4X67A12088 ) ) ) ) ) ) ) ) 2 2 2 2 1 1 1 1

n n n n n n n n e e e e e e e e G G G G G G G G ( ( ( ( ( ( ( (

5 0 0 0 0 0 0 0 0 5 3 5 7 6 3 5 7 6 6 5 6 6 8 5 6 6 8 R D R R R D R R R

Operating systems S S S S S S S S S Microsoft Windows Server 2012 R2 N N N N N Y Y N Y Microsoft Windows Server 2016 Y Y Y N Y Y Y N Y Microsoft Windows Server 2019 Y Y Y N Y Y Y N Y Microsoft Windows Server version 1803 N N N N N N Y N N Red Hat Enterprise Linux 6.10 N N N N N Y Y N Y Red Hat Enterprise Linux 6.9 N N N N N Y Y N Y Red Hat Enterprise Linux 7.3 N N N N N Y Y N N Red Hat Enterprise Linux 7.4 N N N N N Y Y N Y Red Hat Enterprise Linux 7.5 N N N Y N Y Y Y Y Red Hat Enterprise Linux 7.6 Y Y Y Y Y Y Y Y Y Red Hat Enterprise Linux 7.7 Y Y Y Y Y Y Y Y Y Red Hat Enterprise Linux 7.8 Y Y Y Y Y Y Y Y Y Red Hat Enterprise Linux 8.0 Y Y Y Y Y Y Y Y Y Red Hat Enterprise Linux 8.1 Y Y Y Y Y Y Y Y Y Red Hat Enterprise Linux 8.2 Y Y Y Y Y Y Y Y Y SUSE Linux Enterprise Server 11 SP4 N N N N N Y Y N Y SUSE Linux Enterprise Server 11 SP4 with Xen N N N N N Y Y N Y SUSE Linux Enterprise Server 12 SP2 N N N N N Y Y N N SUSE Linux Enterprise Server 12 SP2 with Xen N N N N N Y Y N N SUSE Linux Enterprise Server 12 SP3 N N N N N Y Y N Y SUSE Linux Enterprise Server 12 SP3 with Xen N N N N N Y Y N Y SUSE Linux Enterprise Server 12 SP4 Y Y Y N Y Y Y N Y SUSE Linux Enterprise Server 12 SP4 with Xen Y Y Y N Y Y Y N Y SUSE Linux Enterprise Server 15 N Y Y N Y Y Y N Y SUSE Linux Enterprise Server 15 SP1 Y Y Y Y Y Y Y Y Y SUSE Linux Enterprise Server 15 SP1 with Xen Y Y Y Y Y Y Y Y Y SUSE Linux Enterprise Server 15 with Xen N Y Y N Y Y Y N Y VMware vSphere Hypervisor (ESXi) 6.0 U3 N N N N N Y Y N Y VMware vSphere Hypervisor (ESXi) 6.5 N N N N N Y Y N N VMware vSphere Hypervisor (ESXi) 6.5 U1 N N N N N Y Y N Y VMware vSphere Hypervisor (ESXi) 6.5 U2 N Y Y N Y Y Y N Y

ThinkSystem and ThinkAgile GPU Summary 23 ) ) ) ) ) ) ) ) 2 2 2 2 1 1 1 1

n n n n n n n n e e e e e e e e G G G G G G G G ( ( ( ( ( ( ( (

5 0 0 0 0 0 0 0 0 5 3 5 7 6 3 5 7 6 6 5 6 6 8 5 6 6 8 R D R R R D R R R

Operating systems S S S S S S S S S VMware vSphere Hypervisor (ESXi) 6.5 U3 Y Y Y N Y Y Y N Y VMware vSphere Hypervisor (ESXi) 6.7 N N N N N Y Y N Y VMware vSphere Hypervisor (ESXi) 6.7 U1 N Y Y N Y Y Y N Y VMware vSphere Hypervisor (ESXi) 6.7 U2 N Y Y N Y Y Y N Y VMware vSphere Hypervisor (ESXi) 6.7 U3 Y Y Y Y Y Y Y Y Y VMware vSphere Hypervisor (ESXi) 7.0 Y Y Y Y Y Y Y Y Y

Auxiliary power cables The V100 option part numbers ship with three auxiliary power cables. These cables are needed when the GPU is installed in specific servers, as listed in the following table. For use in the SR670 and SR665, cables ordered with the server are used instead of the cables included with the GPU.

Table 24. Auxiliary power cables for V100 (click images to show larger versions)

Auxiliary power cables included with the GPU option part number 190mm 8pin (2x4) cable Server support: SR860 SBB: SBB7A09174 Base: SC17A31157 FRU: 01GT981

180mm 8pin (2x4) cable Server support: SD530 SBB: SBB7A09053 Base: SC17A07129 FRU: 00MW584

300mm 8pin (2x4) cable Server support: SR650† SBB: SBB7A00299 Base: SC17A02296 FRU: 01KN066

Auxiliary power cables supplied with the SR655 (configure-to-order only) 350mm 8pin (2x4) cable Server support: SR655 (Riser 1 or Riser 2) Feature: B5T5 SBB: SBB7A14640 Base: SC17A50848 FRU: 02JK011

ThinkSystem and ThinkAgile GPU Summary 24 250mm 8pin (2x4) cable Server support: SR655 (Riser 3) Feature: B5TS SBB: SBB7A10974 Base: SC17A50844 FRU: 02JK010

Auxiliary power cables supplied with the SR670 (configure-to-order) 900mm SR670 Cage 1 power cable Feature: B3Y3 SBB: SBB7A10375 Base: SC17A10876 FRU: 01PG448

Notes:

Feature & SBB also include PCIe data cable Cable also supplies power to the riser

1100mm SR670 Cage 2 power cable Feature: B3Y2 SBB: SBB7A10374 Base: SC17A10863 FRU: 01PG426

Notes:

Feature & SBB also include PCIe data cable Cable also supplies power to the riser

Auxiliary power cables needed with the SR665 (field upgrade or configure-to-order) 360mm 8pin (2x4) cable Option: 4M17A11759, ThinkSystem 2U GPU Thermal Option Kit Feature: BAD8 SBB: SBB7A21691 Base: SC17A59596 FRU: 02YE420

Note: The option part number includes other components needed to install the GPU. See theS R665 product guide for details.

† For customers who buy the GPU directly from NVIDIA, this cable is also included in the ThinkSystem SR650 GPU Cable Kit, 4XH7A08794

ThinkSystem and ThinkAgile GPU Summary 25 ThinkSystem NVIDIA Tesla V100 FHHL GPU The NVIDIA Tesla V100 FHHL GPU Accelerator is the latest NVIDIA Volta family product, targets for advanced data center to accelerate AI, HPC, and graphics. Tesla V100 FHHL offers significant performance and great power efficiency. The GPU is a full-height half-length (FHHL) form factor. Introduction Part numbers Technical specifications Server support Operating system support Auxiliary power cables

Introduction The NVIDIA Tesla V100 FHHL GPU Accelerator for PCIe is a single-slot, full-height half-length (FHHL) PCI Express Gen 3 card with NVIDIA Volta graphics processing unit, 16GB HBM2 memory with native ECC support. It uses a passive heat sink for cooling. The Tesla V100 FHHL PCIe supports double precision (FP64), single precision (FP32) and half precision (FP16) compute tasks, unified virtual memory and page migration engine.

Figure 6. NVIDIA Tesla V100 FHHL GPU

Part numbers

Table 25. Ordering information

Part number Feature code Description 4X67A11524 B32D ThinkSystem NVIDIA Tesla V100 FHHL GPU

ThinkSystem and ThinkAgile GPU Summary 26 Technical specifications The following table lists the specifications of the NVIDIA Tesla V100 FHHL GPU.

Table 26. V100 FHHL specifications

Feature Specification GPU Architecture NVIDIA Volta NVIDIA Tensor Cores 640 NVIDIA CUDA® Cores 5,120 Double-Precision Performance 6.5 TFLOPS Single-Precision Performance 13 TFLOPS Tensor Performance 105 TFLOPS GPU Memory 16 GB HBM2 Memory Bandwidth 900 GB/sec ECC Yes Interconnect Bandwidth 32 GB/sec System Interface PCIe Gen3 Form Factor PCIe Full Height/Half Length Max Power Consumption 150 W Thermal Solution Passive Compute APIs CUDA, DirectCompute, OpenCL, OpenACC

Server support The following table lists the ThinkSystem servers that are compatible. The numbers listed in the server columns represent the number of GPUs supported.

Table 27. ThinkSystem server support

Dense/ E 1S Intel 2S Intel AMD 4S Intel Blade ) ) 3 ) 0 ) 3 ) / G 6 ) ) ) ) ) ) ) ) ) ) ) ) ) ) V ) ) 2 2 X Z 8 3 2 8 4 3 9 2 6 9 9 0 X 1 2 6 0 ) 3 2 7 3 1 D / 5 0 0 0 9 0 0 9 1 7 1 0 / 4 1 / / 0 D 2 D 1 9 Y X X Y X X X 7 Y Z X X 2 5 / D Y X 7 7 / 3 5 7 7 7 7 7 7 7 3 7 7 7 7 1 7 7 7 Y / F ) / / / / / / / / / / / / / ) ) ) ) / / / D Z 7 2 Y W 4 1 7 3 2 8 1 5 6 8 0 8 9 1 1 8 6 5 6 / 5 9 7 7 2 2 5 5 0 0 0 9 0 0 3 9 0 1 D 6 1 2 5 1 1 ( ( 4 8 4 0

7 4 Y Y X X Y X X X Y Y Y D D X X X X X X X Z Y X ( 2 2

7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 Y 7 7 ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( V V ( ( ( ( ( ( ( P

7

( 0 0 0 0 0 0 0 0 0 5 5 5 5 0 0 0 0 0 0 0 0 0 0 0 0 0 5 5 3 5 7 9 3 5 7 3 5 4 6 5 5 6 5 5 6 3 5 5 5 5 0 5 5 1 2 5 5 5 5 6 6 6 6 6 6 6 8 8 8 9 8 8 5 6 5 8 3 5 2 5 E T T R R T R R R R R R R R R R R R R R R R R D D N N

Description & Part number S S S S S S S S S S S S S S S S S S S S S S S S S S S NVIDIA Tesla V100 FHHL, N N N N N N N N N N N 3 N N N N N N N N N N N N N N N 4X67A11524

ThinkSystem and ThinkAgile GPU Summary 27 Operating system support The following table lists the supported operating systems.

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

Table 28. Operating system support for ThinkSystem NVIDIA Tesla V100 FHHL 16GB PCIe Passive GPU , 4X67A11524 ) ) 2 1

n n e e G G ( (

0 0 5 5 6 6 R R

Operating systems S S Microsoft Windows Server 2012 R2 N Y Microsoft Windows Server 2016 Y Y Microsoft Windows Server 2019 Y Y Red Hat Enterprise Linux 6.10 N Y Red Hat Enterprise Linux 6.9 N Y Red Hat Enterprise Linux 7.3 N Y Red Hat Enterprise Linux 7.4 N Y Red Hat Enterprise Linux 7.5 N Y Red Hat Enterprise Linux 7.6 Y Y Red Hat Enterprise Linux 7.7 Y Y Red Hat Enterprise Linux 7.8 Y Y Red Hat Enterprise Linux 8.0 Y Y Red Hat Enterprise Linux 8.1 Y Y Red Hat Enterprise Linux 8.2 Y Y SUSE Linux Enterprise Server 11 SP4 N Y SUSE Linux Enterprise Server 11 SP4 with Xen N Y SUSE Linux Enterprise Server 12 SP2 N Y SUSE Linux Enterprise Server 12 SP2 with Xen N Y SUSE Linux Enterprise Server 12 SP3 N Y SUSE Linux Enterprise Server 12 SP3 with Xen N Y SUSE Linux Enterprise Server 12 SP4 Y Y SUSE Linux Enterprise Server 12 SP4 with Xen Y Y SUSE Linux Enterprise Server 15 Y Y SUSE Linux Enterprise Server 15 SP1 Y Y SUSE Linux Enterprise Server 15 SP1 with Xen Y Y SUSE Linux Enterprise Server 15 with Xen Y Y VMware vSphere Hypervisor (ESXi) 6.5 N Y VMware vSphere Hypervisor (ESXi) 6.5 U1 N Y VMware vSphere Hypervisor (ESXi) 6.5 U2 Y Y VMware vSphere Hypervisor (ESXi) 6.5 U3 Y Y VMware vSphere Hypervisor (ESXi) 7.0 Y Y

ThinkSystem and ThinkAgile GPU Summary 28 Auxiliary power cables The V100 FHHL option part number ships with one auxiliary power cable. The cable is needed when the GPU is installed in specific servers, as listed in the following table.

Table 29. Auxiliary power cables (click images to show larger versions)

Auxiliary power cables included with the V100 FHHL option part number 300mm 2x 8pin (2x4) cable Server support: SR650† SBB: SBB7A05971 Base: SC17A30059 FRU: 01KR723

† For customers who buy the GPU directly from NVIDIA, this cable is also included in the ThinkSystem SR650 GPU Cable Kit, 4XH7A08794

ThinkSystem and ThinkAgile GPU Summary 29 ThinkSystem NVIDIA Tesla T4 GPU The NVIDIA Tesla T4 GPU supports diverse cloud workloads, including high-performance computing, deep learning training and inference, machine learning, data analytics, and graphics. Based on the new NVIDIA Turing Architecture and packaged in an energy-efficient 70-watt, small PCIe form factor, Tesla T4 is optimized for scale-out computing environments with its multi-precision Turing Tensor Cores and new RT Cores. Introduction Part numbers Technical specifications Server support Operating system support Auxiliary power cables

Introduction The NVIDIA Tesla T4 GPU is a single-slot, low-profile, 6.6-inch PCIe 3.0 Universal Deep Learning Accelerator. The Tesla T4 has 16GB GDDR6 memory. Tesla T4 provides revolutionary multi-precision performance to accelerate deep learning and machine learning training and inference, video transcoding and virtual desktops. Powering performance from FP32 to FP16 to INT8, as well as INT4 precisions, T4 delivers up to 40x higher performance than CPUs. As part of the NVIDIA AI platform, Tesla T4 supports all AI frameworks and network types, delivering dramatic performance and efficiency.

Figure 7. ThinkSystem NVIDIA Tesla T4 GPU

Part numbers

Table 30. Ordering information

Part number Feature code Description 4X67A14926 B4YB ThinkSystem NVIDIA Tesla T4 16GB PCIe Passive GPU

ThinkSystem and ThinkAgile GPU Summary 30 Technical specifications The following table lists the specifications of the NVIDIA Tesla T4 GPU.

Table 31. Technical Specifications

Feature Specification GPU Architecture NVIDIA Turing NVIDIA Turing Tensor Cores 320 cores NVIDIA CUDA Cores 2560 cores Single Precision Performance (FP32) 8.1 TFLOPS Mixed Precision (FP16/FP32) 65 FP16 TFLOPS INT8 Precision 130 INT8 TOPS INT4 Precision 260 INT4 TOPS System Interface PCIe 3.0 x16 or x8 (x16 physical connector) GPU Memory 16 GB GDDR6 Memory Bandwidth 320+ GB/s Form Factor single-slot, low-profile Max Power Consumption 70 W Thermal Solution Passive Compute APIs CUDA, DirectCompute, OpenCL, OpenACC

Server support The following table lists the ThinkSystem servers that are compatible. The numbers listed in the server columns represent the number of GPUs supported.

Table 32. ThinkSystem server support

Dense/ E 1S Intel 2S Intel AMD 4S Intel Blade ) ) 3 ) 0 ) 3 ) / G 6 ) ) ) ) ) ) ) ) ) ) ) ) ) ) V ) ) 2 2 X Z 8 3 2 8 4 3 9 2 6 9 9 0 X 1 2 6 0 ) 3 2 7 3 1 D / 5 0 0 0 9 0 0 9 1 7 1 0 / 4 1 / / 0 D 2 D 1 9 Y X X Y X X X 7 Y Z X X 2 5 / Y X D 7 7 / 3 5 7 7 7 7 7 7 7 3 7 7 7 7 1 7 7 7 Y / F ) / / / / / / / / / / / / / ) ) ) ) / / / D Z 7 2 Y W 4 1 7 3 2 8 1 5 6 8 0 8 9 1 1 8 6 5 5 9 6 / 7 7 2 2 5 5 0 0 0 9 0 0 3 9 0 1 6 1 2 5 1 1 D ( ( 4 0 4 8

7 4 Y Y X X Y X X X Y Y Y D D X X X X X X X Z Y X ( 2 2

7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 Y 7 7 ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( V V ( ( ( ( ( ( ( P

7

( 0 0 0 0 0 0 0 0 0 5 5 5 5 0 0 0 0 0 0 0 0 0 0 0 0 0 5 5 3 5 7 9 3 5 7 3 5 4 6 5 5 6 5 5 6 3 5 5 5 5 0 5 5 1 2 5 5 5 5 6 6 6 6 6 6 6 8 8 8 9 8 8 5 6 5 8 3 5 2 5 E T T R R T R R R R R R R R R R R R R R R R R D D N N

Description & Part number S S S S S S S S S S S S S S S S S S S S S S S S S S S NVIDIA Tesla T4 16GB, 1 N N N N N N N N N 2 5* 8 3 6 3 8 N N N N 2 8 N N N N 4X67A14926 * The SR650 has support for 5x T4 or 5x P4 GPUs in servers with second-generation Intel Xeon Scalable processors only. SR650 systems originally with first-generation processors have support for up to 3x T4 or 2x P4 GPUs.

Operating system support The following table lists the supported operating systems.

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

ThinkSystem and ThinkAgile GPU Summary 31 Table 33. Operating system support for ThinkSystem NVIDIA Tesla T4 16GB PCIe Passive GPU, 4X67A14926 ) ) ) ) ) ) 2 2 2 1 1 1

n n n n n n e e e e e e 2 2 G G G G G G ( ( ( V V ( ( (

5 5 5 5 0 0 0 0 0 0 0 0 0 3 4 5 6 3 5 7 5 6 3 5 7 5 6 6 6 6 6 6 6 8 8 6 6 6 3 E R R R R R R R R R R R R

Operating systems S S S S S S S S S S S S S Microsoft Windows Server 2012 R2 N N N N N N N N N N Y Y N Microsoft Windows Server 2016 Y Y Y Y Y Y Y N Y Y Y Y N Microsoft Windows Server 2019 Y Y Y Y Y Y Y N Y Y Y Y N Red Hat Enterprise Linux 6.10 N N N N N N N N N N Y Y N Red Hat Enterprise Linux 6.9 N N N N N N N N N N Y Y N Red Hat Enterprise Linux 7.3 N N N N N N N N N N Y Y N Red Hat Enterprise Linux 7.4 N N N N N N N N N N Y Y N Red Hat Enterprise Linux 7.5 N N N N N N N Y N N Y Y Y Red Hat Enterprise Linux 7.6 Y Y N Y N Y Y Y N N Y Y Y Red Hat Enterprise Linux 7.7 Y Y N Y N Y Y Y N N Y Y Y Red Hat Enterprise Linux 7.8 Y Y N Y N Y Y Y N N Y Y Y Red Hat Enterprise Linux 7.9 N N N N N N N N Y Y N N N Red Hat Enterprise Linux 8.0 N Y N Y N Y Y Y N N Y Y Y Red Hat Enterprise Linux 8.1 Y Y N Y N Y Y Y N N Y Y Y Red Hat Enterprise Linux 8.2 Y Y N Y N Y Y Y Y Y Y Y Y SUSE Linux Enterprise Server 11 SP4 N N N N N N N N N N Y Y N SUSE Linux Enterprise Server 11 SP4 with Xen N N N N N N N N N N Y Y N SUSE Linux Enterprise Server 12 SP2 N N N N N N N N N N Y Y N SUSE Linux Enterprise Server 12 SP2 with Xen N N N N N N N N N N Y Y N SUSE Linux Enterprise Server 12 SP3 N N N N N N N N N N Y Y N SUSE Linux Enterprise Server 12 SP3 with Xen N N N N N N N N N N Y Y N SUSE Linux Enterprise Server 12 SP4 N Y N Y N Y Y N N N Y Y N SUSE Linux Enterprise Server 12 SP4 with Xen N Y N Y N Y Y N N N Y Y N SUSE Linux Enterprise Server 12 SP5 N N N N N N N N Y Y N N N SUSE Linux Enterprise Server 12 SP5 with Xen N N N N N N N N Y Y N N N SUSE Linux Enterprise Server 15 Y N N N N Y Y N N N Y Y N SUSE Linux Enterprise Server 15 SP1 N Y Y Y Y Y Y Y N N Y Y Y SUSE Linux Enterprise Server 15 SP1 with Xen N Y Y Y Y Y Y Y N N Y Y Y SUSE Linux Enterprise Server 15 SP2 N N N N N N N N Y Y N N N SUSE Linux Enterprise Server 15 SP2 with Xen N N N N N N N N Y Y N N N SUSE Linux Enterprise Server 15 with Xen Y N N N N Y Y N N N Y Y N VMware vSphere Hypervisor (ESXi) 6.5 N N N N N N N N N N Y Y N VMware vSphere Hypervisor (ESXi) 6.5 U1 N N N N N N N N N N Y Y N VMware vSphere Hypervisor (ESXi) 6.5 U2 Y N N N N Y Y N N N Y Y N VMware vSphere Hypervisor (ESXi) 6.5 U3 N Y N Y N Y Y N N N Y Y N

ThinkSystem and ThinkAgile GPU Summary 32 ) ) ) ) ) ) 2 2 2 1 1 1

n n n n n n e e e e e e 2 2 G G G G G G ( ( ( V V ( ( (

5 5 5 5 0 0 0 0 0 0 0 0 0 3 4 5 6 3 5 7 5 6 3 5 7 5 6 6 6 6 6 6 6 8 8 6 6 6 3 E R R R R R R R R R R R R

Operating systems S S S S S S S S S S S S S VMware vSphere Hypervisor (ESXi) 6.7 N N N N N N N N N N Y Y N VMware vSphere Hypervisor (ESXi) 6.7 U1 N N N N N Y Y N N N Y Y N VMware vSphere Hypervisor (ESXi) 6.7 U2 Y N N N N Y Y N N N Y Y N VMware vSphere Hypervisor (ESXi) 6.7 U3 N Y Y Y Y Y Y Y N N Y Y Y VMware vSphere Hypervisor (ESXi) 7.0 Y Y N Y N Y Y Y N N Y Y Y

Auxiliary power cables The T4 does not require an auxiliary power cable.

ThinkSystem and ThinkAgile GPU Summary 33 ThinkSystem NVIDIA Tesla M10 GPU ThinkSystem NVIDIA Tesla M10 GPU accelerator works with NVIDIA GRID software to provide the industry’s highest user density for virtualized desktops and applications. It supports 64 desktops per board and 128 desktops per server, giving your business the power to deliver great experiences to all of your employees at an affordable cost. Introduction Part numbers Technical specifications Server support Operating system support Auxiliary power cables

Introduction ThinkSystem NVIDIA Tesla M10 is a dual-slot 10.5 inch PCI Express 3.0 graphics card with four mid-range NVIDIA Maxwell graphics processing units (GPUs). The Tesla M10 has 32 GB GDDR5 memory (8 GB per GPU) and a 225 W maximum power limit. The board is passively cooled and supports both airflow directions.

Figure 8. ThinkSystem NVIDIA Tesla M10 GPU

Part numbers

Table 34. Ordering information

Part number Feature code Description 7C57A02891 AX8L, B15V, B0RK* ThinkSystem NVIDIA Tesla M10 * Feature code varies by system

ThinkSystem and ThinkAgile GPU Summary 34 Technical specifications The following table lists the specifications of the NVIDIA Tesla M10 GPU.

Table 35. M10 specifications

Feature Specification Virtualization Use Case Performance-Optimized Graphics Virtualization GPU Architecture NVIDIA Maxwell GPUs per Board 4 Max User per Board 64 (16 per GPU) NVIDIA CUDA® Cores 2560 NVIDIA CUDA Cores (640 per GPU) GPU Memory 32 GB of GDDR5 Memory (8 per GPU) H.264 1080p30 streams 28 Max Power Consumption 225 W Thermal Solution Active/Passive Passive Form Factor PCIe 3.0 Dual Slot PCIe 3.0 Dual Slot

Server support The following table lists the ThinkSystem servers that are compatible. The numbers listed in the server columns represent the number of GPUs supported.

Server memory limitation: When the M10 is installed, the total system memory must be less than 1 TB. See https://support.lenovo.com/us/en/solutions/ht114952 for details.

Table 36. ThinkSystem server support

Dense/ E 1S Intel 2S Intel AMD 4S Intel Blade ) ) 3 ) 0 ) 3 ) / G 6 ) ) ) ) ) ) ) ) ) ) ) ) ) ) V ) ) 2 2 X Z 8 3 2 8 4 3 9 2 6 9 9 0 X 1 2 ) 6 0 3 2 7 3 1 D / 5 0 0 0 9 0 0 9 1 7 1 0 / 0 4 1 / / D 2 D 1 9 5 Y X X Y X X X 7 Y Z X X 2 / D Y X 7 7 / 3 5 7 7 7 7 7 7 7 3 7 7 7 7 1 7 Y 7 7 / F ) / / / / / / / / / / / / / ) ) ) ) / / / D Z 7 2 Y W 4 1 7 3 2 8 1 5 6 8 0 8 9 1 1 8 6 5 6 / 5 9 7 7 2 2 5 5 0 0 0 9 0 0 3 9 0 1 D 6 1 2 5 1 1 ( ( 4 8 4 0

7 4 Y Y X X Y X X X Y Y Y D D X X X X X X X Z Y X ( 2 2

7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 Y 7 7 ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( V V ( ( ( ( ( ( ( P

7

( 0 0 0 0 0 0 0 0 0 5 5 5 5 0 0 0 0 0 0 0 0 0 0 0 0 0 5 5 3 5 7 9 3 5 7 3 5 4 6 5 5 6 5 5 6 3 5 5 5 5 0 5 5 1 2 5 5 5 5 6 6 6 6 6 6 6 8 8 8 9 8 8 5 6 5 8 3 5 2 5 E T T R R T R R R R R R R R R R R R R R R R R D D N N

Description & Part number S S S S S S S S S S S S S S S S S S S S S S S S S S S NVIDIA Tesla M10, N N N N N N N N N N N 2 N N N N N N N N N N N 2 N N N 7C57A02891

ThinkSystem and ThinkAgile GPU Summary 35 Operating system support The following table lists the supported operating systems.

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

Table 37. Operating system support for NVIDIA Tesla M10 32GB GPU, PCIe (Passive), 7C57A02891 ) ) 5 1 ) ) ) ) ) 6 7 2 2 2 1 1 4 8 3

5 8 5 n n n n ( (

9

e e e e ( 5 5 G G G G 5 ( ( ( ( M M

M 0 0 0 0 0 0 3 5 3 5 6 5 0 5 6 5 6 3 6 4 D R D R x 3 2

Operating systems S S S S n x x Microsoft Windows Server 2012 N N N N Y Y N Microsoft Windows Server 2012 R2 N N N N Y Y N Microsoft Windows Server 2016 Y Y Y Y Y Y N Microsoft Windows Server 2019 Y Y Y Y N N N VMware vSphere Hypervisor (ESXi) 6.0 U3 N N Y Y Y Y N VMware vSphere Hypervisor (ESXi) 6.5 N N Y Y Y Y N VMware vSphere Hypervisor (ESXi) 6.5 U1 N N Y Y Y Y N VMware vSphere Hypervisor (ESXi) 6.5 U2 Y Y Y Y Y Y N VMware vSphere Hypervisor (ESXi) 6.5 U3 Y Y Y Y Y Y N VMware vSphere Hypervisor (ESXi) 6.7 N N Y Y Y Y Y VMware vSphere Hypervisor (ESXi) 6.7 U1 Y Y Y Y Y Y Y VMware vSphere Hypervisor (ESXi) 6.7 U2 Y Y Y Y Y Y Y VMware vSphere Hypervisor (ESXi) 6.7 U3 Y Y Y Y Y Y Y VMware vSphere Hypervisor (ESXi) 7.0 Y Y Y Y N N N

ThinkSystem and ThinkAgile GPU Summary 36 Auxiliary power cables The M10 option part number ships with two auxiliary power cables. These cables are needed when the GPU is installed in specific servers, as listed in the following table.

Table 38. Auxiliary power cables (click images to show larger versions)

Auxiliary power cables included with the M10 option part number 130mm PCIe 6pin (2x3) + PCIe 8pin (2x4): Server support: SD530 SBB: SBB7A09207 Base: SC17A07130 FRU: 00MW585

300mm PCIe 6pin (2x3) + PCIe 8pin (2x4): Server support: SR650† SBB: SBB7A00293 Base: SC17A02243 FRU: 01KN070

† For customers who buy the GPU directly from NVIDIA, this cable is also included in the ThinkSystem SR650 GPU Cable Kit, 4XH7A08794

ThinkSystem and ThinkAgile GPU Summary 37 ThinkSystem AMD Radeon Instinct MI25 GPU The Radeon Instinct MI25 accelerator, designed with the most advanced Next-Gen “Vega” GPU architecture, is the ultimate training accelerator for large scale machine intelligence and deep learning, along with being an optimized open compute workhorse for single-precision HPC-class system workloads. Introduction Part numbers Technical specifications Server support Operating system support Auxiliary power cables

Introduction The MI25 delivers leadership in FP16 and FP32 performance in a passively-cooled single GPU server card with 24.6 TFLOPS of FP16 and 12.3 TFLOPS of FP32 peak performance through its 64 compute units with 4,096 stream processors. The Radeon Instinct MI25’s powerful compute engine and advanced memory architecture, combined with AMD’s ROCm open software platform and ecosystem, provides a powerful, flexible heterogeneous compute solution that allows datacenter designers to meet the challenges of a new era of compute and Machine Intelligence.

Figure 9. AMD Radeon Instinct MI25 GPU

Part numbers

Table 39. Ordering information

Part number Feature code Description 7C57A02897 B228 ThinkSystem AMD Radeon Instinct MI25 16GB PCIe Passive GPU 4C57A16224 B5DN ThinkSystem SR670 AMD Radeon Instinct MI25 16Gb PCIe Passive GPU

ThinkSystem and ThinkAgile GPU Summary 38 Technical specifications The following table lists the specifications of the AMD Radeon Instinct MI25 GPU.

Table 40. Technical specifications

Feature Specification GPU Architecture: AMD “Vega10” Stream Processors: 4096 GPU Memory: 16GB HBM2 Memory Bandwidth: Up to 484 GB/s Half-Precision Performance (FP16) 24.6 TFLOPS Single-Precision Performance (FP32) 12.3 TFLOPS Double-Precision Performance (FP64) 768 GFLOPS ECC: Yes Bus Interface: PCIe Gen 3 x16 MxGPU Capability: Yes Board Form Factor: Full-Height, Dual-Slot Thermal Solution: Passively Cooled Standard Max Power: 300W TDP OS Support: Linux 64-bit ROCm Software Platform: Yes Programing Environment: ISO C++, OpenCL, CUDA (via AMD’s HIP conversion tool) and Python5 (via Anaconda’s NUMBA)

Server support The following table lists the ThinkSystem servers that are compatible. The numbers listed in the server columns represent the number of GPUs supported.

Table 41. ThinkSystem server support

Dense/ E 1S Intel 2S Intel AMD 4S Intel Blade ) ) 3 ) 0 3 ) / ) 6 G ) ) ) ) ) ) ) ) ) ) ) ) ) ) V ) ) 2 2 Z X 8 3 2 8 4 3 9 2 6 9 9 0 X 1 2 ) 6 0 3 7 2 3 1 D / 5 0 0 0 9 0 0 9 1 7 1 0 / 0 4 1 / / D 2 D 1 9 5 Y X X Y X X X 7 Y Z X X 2 / D Y X 7 7 / 3 5 7 7 7 7 7 7 7 3 7 7 7 7 1 7 Y 7 7 / F ) / / / / / / / / / / / / / ) ) ) ) / / / D Z 7 2 Y W 4 1 7 3 2 8 1 5 6 8 0 8 9 1 1 8 6 5 6 / 5 9 7 7 2 2 5 5 0 0 0 9 0 0 3 9 0 1 D 6 1 2 5 1 1 ( ( 4 8 4 0

7 4 Y Y X X Y X X X Y Y Y D D X X X X X X X Z Y X ( 2 2

7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 Y 7 7 ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( V V ( ( ( ( ( ( ( P

7

( 0 0 0 0 0 0 0 0 0 5 5 5 5 0 0 0 0 0 0 0 0 0 0 0 0 0 5 5 3 5 7 9 3 5 7 3 5 4 6 5 5 6 5 5 6 3 5 5 5 5 0 5 5 1 2 5 5 5 5 6 6 6 6 6 6 6 8 8 8 9 8 8 5 6 5 8 3 5 2 5 E T T R R T R R R R R R R R R R R R R R R R R D D N N

Description & Part number S S S S S S S S S S S S S S S S S S S S S S S S S S S AMD Radeon Instinct MI25, N N N N N N N N N N N 2 N N N N N N N N N N N 2 N N N 7C57A02897 AMD Radeon Instinct MI25 N N N N N N N N N N N N 4 N N N N N N N N N N N N N N (SR670), 4C57A16224

ThinkSystem and ThinkAgile GPU Summary 39 Operating system support The following tables list the supported operating systems.

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

Table 42. Operating system support for ThinkSystem AMD Radeon Instinct MI25 16GB PCIe Passive GPU, 7C57A02897 ) ) ) ) 2 2 1 1

n n n n e e e e G G G G ( ( ( (

0 0 0 0 3 5 3 5 5 6 5 6 D R D R

Operating systems S S S S Red Hat Enterprise Linux 7.3 N N Y Y Red Hat Enterprise Linux 7.4 N N Y Y Red Hat Enterprise Linux 7.5 N N Y Y Red Hat Enterprise Linux 7.6 Y Y Y Y

Table 43. Operating system support for ThinkSystem SR670 AMD Radeon Instinct MI25 16Gb PCIe Passive GPU, 4C57A16224 ) ) 2 1

n n e e G G ( (

0 0 7 7 6 6 R R

Operating systems S S Red Hat Enterprise Linux 7.5 Y Y Red Hat Enterprise Linux 7.6 Y Y

ThinkSystem and ThinkAgile GPU Summary 40 Auxiliary power cables The MI25 option part number 7C57A02897 ships with three auxiliary power cables. These cables are needed when the GPU is installed in specific servers, as listed in the following table.

Table 44. Auxiliary power cables included with 7C57A02897 (click images to show larger versions)

Auxiliary power cables included with MI25 option part number 7C57A02897 130mm PCIe 6pin (2x3) + PCIe 8pin (2x4): Server support: SD530 SBB: SBB7A09207 Base: SC17A07130 FRU: 00MW585

300mm PCIe 6pin (2x3) + PCIe 8pin (2x4): Server support: SR650† SBB: SBB7A00293 Base: SC17A02243 FRU: 01KN070

130mm 2x PCIe 8pin (2x4): Server support: SR860* SBB: SBB7A09173 Base: SC17A35678 FRU: 01GT957

* The SR860 server does not support the MI25 GPU even though this cable is included in part 7C57A02897 † For customers who buy the GPU directly from NVIDIA, this cable is also included in the ThinkSystem SR650 GPU Cable Kit, 4XH7A08794

ThinkSystem and ThinkAgile GPU Summary 41 The MI25 option part number for SR670, 4C57A16224, ships with one auxiliary power cable. This cable is needed when the GPU is installed in the SR670 and is used in conjunction with the Cage 1 and Cage 2 power cables that are included with the SR670.

Table 45. Auxiliary power cables included with 4C57A16224 (click images to show larger versions)

Auxiliary power cables included with the MI25 SR670 option part number 4C57A16224 100mm PCIe 6pin (2x3) + PCIe 8pin (2x4): Server support: SR670 Option: 4Z57A26300* SBB: SBB7A16972 Base: SC17A10850 FRU: 01PG437

Note: This cable is used in conjunction with either SBB7A10375 or SBB7A10374.

Auxiliary power cables supplied with the SR670 (configure-to-order) 900mm SR670 Cage 1 power cable Feature: B3Y3 SBB: SBB7A10375 Base: SC17A10876 FRU: 01PG448

Notes:

Feature & SBB also include PCIe data cable Cable also supplies power to the riser

1100mm SR670 Cage 2 power cable Feature: B3Y2 SBB: SBB7A10374 Base: SC17A10863 FRU: 01PG426

Notes:

Feature & SBB also include PCIe data cable Cable also supplies power to the riser

* This cable is also available as a separate option part number for customers that buy the MI25 directly from AMD and wish to use the adapter in the SR670 server.

ThinkSystem and ThinkAgile GPU Summary 42 ThinkSystem NVIDIA Quadro RTX 8000 GPU Bring the power of RTX to the data center with the NVIDIA Quadro RTX 8000, and Quadro Virtual Data Center Workstation (Quadro vDWS) software, built on the NVIDIA Turing architecture and the NVIDIA RTX platform for powerful server-based visual computing solutions. Accelerate multiple data center workloads including batch rendering, data science, virtual workstation, simulation, and augmented or virtual reality over 5G networks. Customers can also serve multiple powerful virtual workstations with Quadro vDWS software. Introduction Part numbers Technical specifications Server support Auxiliary power cables

Introduction NVIDIA Quadro RTX 8000 has 4,608 CUDA cores, 576 Tensor Cores, 72 RT Cores, and 48 GB of high- performance graphics memory, and delivers the best performance and the largest graphics memory for the most demanding visual computing tasks.

Figure 10. ThinkSystem NVIDIA Quadro RTX 8000 Passive GPU

Part numbers

Table 46. Ordering information

Part number Feature code Description 4X67A65441* BD4R, BCGR ThinkSystem NVIDIA Quadro RTX 8000 48GB PCIe Passive GPU

ThinkSystem and ThinkAgile GPU Summary 43 * The NVIDIA RTX 8000 is available via Lenovo Scalable Infrastructure, LeSI. Select "AI & HPC – LeSI Solutions" in the DCSC configurator. See the LeSI product guide for details.

Technical specifications The following table lists the specifications of the NVIDIA Quadro RTX 8000 GPU.

Table 47. RTX 8000 specifications

Feature Specification GPU Memory 48 GB GDDR6 Memory Interface 384-bit Memory Bandwidth Up to 624 GB/s ECC Yes NVIDIA CUDA Cores 4,608 NVIDIA Tensor Cores 576 NVIDIA RT Cores 72 Single-Precision Performance 14.9 TFLOPS Tensor Performance 119.4 TFLOPS Host Interface PCI Express 3.0 x 16 Power Consumption 250 W Thermal Solution Passive cooling Form Factor 4.4” H x 10.5” L, Dual Slot, Full Height, Full Length Display Connectors None Encode / Decode Engines 1X Encode, 1X Decode VR Ready Yes Graphics APIs Shader Model 5.1, OpenGL 4.5, DirectX 12 Compute APIs CUDA, DirectCompute, OpenCL, OpenACC

Server support The following table lists the ThinkSystem servers that are compatible. The numbers listed in the server columns represent the number of GPUs supported.

Table 48. ThinkSystem server support

Dense/ E 1S Intel 2S Intel AMD 4S Intel Blade ) ) 3 ) 0 3 ) / ) 6 G ) ) ) ) ) ) ) ) ) ) ) ) ) ) V ) ) 2 2 Z X 8 3 2 8 4 3 9 2 6 9 9 0 X 1 2 ) 6 0 3 7 2 3 1 D / 5 0 0 0 9 0 0 9 1 7 1 0 / / / 0 4 1 D 2 D 1 9 Y X X Y X X X 7 Y Z X X 2 5 / D Y X 7 7 / 3 5 7 7 7 7 7 7 7 3 7 7 7 7 1 7 Y 7 7 / F ) / / / / / / / / / / / / / ) ) ) ) / / / D Z 7 2 Y W 4 1 7 3 2 8 1 5 6 8 0 8 9 1 1 8 6 5 6 / 5 9 7 7 2 2 5 5 0 0 0 9 0 0 3 9 0 1 D 6 1 2 5 1 1 ( ( 4 8 4 0

7 Y Y X X Y X X X Y Y Y D D X X X X X X X 4 Z Y X ( 2 2

7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 Y 7 7 ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( V V ( ( ( ( ( ( ( P

7

( 0 0 0 0 0 0 0 0 0 5 5 5 5 0 0 0 0 0 0 0 0 0 0 0 0 0 5 5 3 5 7 9 3 5 7 3 5 4 6 5 5 6 5 5 6 3 5 5 5 5 0 5 5 1 2 5 5 5 5 6 6 6 6 6 6 6 8 8 8 9 8 8 5 6 5 8 3 Description & Part 5 2 5 E T T R R T R R R R R R R R R R R R R R R R R D D N N

number S S S S S S S S S S S S S S S S S S S S S S S S S S S NVIDIA Quadro RTX 8000, N N N N N N N N N N N 2† 4† N 3† N N N N N N N N N N N N 4X67A65441

ThinkSystem and ThinkAgile GPU Summary 44 † Only available via Lenovo Scalable Infrastructure (LeSI). Select "AI & HPC – LeSI Solutions" in the DCSC configurator. See the LeSI product guide for details.

Auxiliary power cables The RTX 8000 option part number ships with one auxiliary power cables. This cable is needed when the GPU is installed in specific servers, as listed in the following table. For use in the SR670, cables ordered with the server are used instead of the cables included with the GPU.

Table 49. Auxiliary power cables for RTX 8000 (click images to show larger versions)

Auxiliary power cable included with the GPU option part number 300mm 8pin (2x4) cable Server support: SR650†, SR655 SBB: SBB7A00299 Base: SC17A02296 FRU: 01KN066

Auxiliary power cables supplied with the SR670 (configure-to-order) 900mm SR670 Cage 1 power cable Feature: B3Y3 SBB: SBB7A10375 Base: SC17A10876 FRU: 01PG448

Notes:

Feature & SBB also include PCIe data cable Cable also supplies power to the riser

1100mm SR670 Cage 2 power cable Feature: B3Y2 SBB: SBB7A10374 Base: SC17A10863 FRU: 01PG426

Notes:

Feature & SBB also include PCIe data cable Cable also supplies power to the riser

† For customers who buy the GPU directly from NVIDIA, this cable is also included in the ThinkSystem SR650 GPU Cable Kit, 4XH7A08794

ThinkSystem and ThinkAgile GPU Summary 45 ThinkSystem NVIDIA Quadro RTX 6000 GPU Bring the power of RTX to the data center with the NVIDIA Quadro RTX 6000 and Quadro Virtual Data Center Workstation (Quadro vDWS) software for powerful server-based visual computing solutions. The NVIDIA Quadro RTX 6000 is ideal for high-performance server-based visual computing workloads including rendering, data science, virtual workstation, simulation, and scientific visualization. The RTX 6000 is optimized for reliability in enterprise data centers and built for 24/7 server environments. Introduction Part numbers Technical specifications Server support Operating system support Auxiliary power cables

Introduction The Quadro RTX 6000 is equipped with 4,608 CUDA cores, 576 Tensor cores, 72 RT cores and 24GB GDDR6 memory. It can render complex models and scenes with physically accurate shadows, reflections, and refractions to empower users with instant insight.

Figure 11. ThinkSystem NVIDIA Quadro RTX 6000 Passive GPU

Part numbers

Table 50. Ordering information

Part number Feature code Description 4X67A13125 BB2D ThinkSystem NVIDIA Quadro RTX 6000 24GB PCIe Passive GPU

ThinkSystem and ThinkAgile GPU Summary 46 Technical specifications The following table lists the specifications of the NVIDIA Quadro RTX 6000 GPU.

Table 51. RTX 6000 specifications

Feature Specification GPU Memory 24 GB GDDR6 Memory Interface 384-bit Memory Bandwidth Up to 624 GB/s ECC Yes NVIDIA CUDA Cores 4,608 NVIDIA Tensor Cores 576 NVIDIA RT Cores 72 Single-Precision Performance 14.9 TFLOPS Tensor Performance 119.4 TFLOPS Host Interface PCI Express 3.0 x 16 Power Consumption 250 W Thermal Solution Passive cooling Form Factor 4.4” H x 10.5” L, Dual Slot, Full Height, Full Length Display Connectors None Encode / Decode Engines 1X Encode, 1X Decode VR Ready Yes Graphics APIs DirectX 12.0, Shader Model 5.1, OpenGL 4.5, Vulkan 1.0 Compute APIs CUDA, DirectCompute, OpenCL

Server support The following table lists the ThinkSystem servers that are compatible. The numbers listed in the server columns represent the number of GPUs supported.

Table 52. ThinkSystem server support

Dense/ E 1S Intel 2S Intel AMD 4S Intel Blade ) ) 3 ) 0 3 ) / ) 6 G ) ) ) ) ) ) ) ) ) ) ) ) ) ) V ) ) 2 2 Z X 8 3 2 8 4 3 9 2 6 9 9 0 X 1 2 ) 6 0 3 7 2 3 1 D / 5 0 0 0 9 0 0 9 1 7 1 0 / 0 4 1 / / D 2 D 1 9 5 Y X X Y X X X 7 Y Z X X 2 / D Y X 7 7 / 3 5 7 7 7 7 7 7 7 3 7 7 7 7 1 7 Y 7 7 / F ) / / / / / / / / / / / / / ) ) ) ) / / / D Z 7 2 Y W 4 1 7 3 2 8 1 5 6 8 0 8 9 1 1 8 6 5 6 / 5 9 7 7 2 2 5 5 0 0 0 9 0 0 3 9 0 1 D 6 1 2 5 1 1 ( ( 4 8 4 0

7 4 Y Y X X Y X X X Y Y Y D D X X X X X X X Z Y X ( 2 2

7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 Y 7 7 ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( V V ( ( ( ( ( ( ( P

7

( 0 0 0 0 0 0 0 0 0 5 5 5 5 0 0 0 0 0 0 0 0 0 0 0 0 0 5 5 3 5 7 9 3 5 7 3 5 4 6 5 5 6 5 5 6 3 5 5 5 5 0 5 5 1 2 5 5 5 5 6 6 6 6 6 6 6 8 8 8 9 8 8 5 6 5 8 3 5 2 5 E T T R R T R R R R R R R R R R R R R R R R R D D N N

Description & Part number S S S S S S S S S S S S S S S S S S S S S S S S S S S NVIDIA Quadro RTX 6000, N N N N N N N N N N N 2† 4 N 3† N N N N N N N N N N N N 4X67A13125 † Only available via Lenovo Scalable Infrastructure (LeSI). Select "AI & HPC – LeSI Solutions" in the DCSC configurator. See the LeSI product guide for details.

ThinkSystem and ThinkAgile GPU Summary 47 Operating system support The following table lists the supported operating systems.

Tip: This table is automatically generated based on data from Lenovo ServerProven.

Table 53. Operating system support for ThinkSystem NVIDIA Quadro RTX 6000 24GB PCIe Passive, 4X67A13125 ) ) 2 1

n n e e G G ( (

0 0 7 7 6 6 R R

Operating systems S S Red Hat Enterprise Linux 7.5 Y Y Red Hat Enterprise Linux 7.6 Y Y Red Hat Enterprise Linux 7.7 Y Y Red Hat Enterprise Linux 7.8 Y Y Red Hat Enterprise Linux 8.0 Y Y Red Hat Enterprise Linux 8.1 Y Y Red Hat Enterprise Linux 8.2 Y Y SUSE Linux Enterprise Server 15 SP1 Y Y SUSE Linux Enterprise Server 15 SP1 with Xen Y Y VMware vSphere Hypervisor (ESXi) 6.7 U3 Y Y VMware vSphere Hypervisor (ESXi) 7.0 Y Y

ThinkSystem and ThinkAgile GPU Summary 48 Auxiliary power cables The RTX 6000 option part number ships with one auxiliary power cable. This cable is needed when the GPU is installed in the SR670 and is used in conjunction with the Cage 1 and Cage 2 power cables that are included with the SR670.

Table 54. Auxiliary power cables (click images to show larger versions)

Auxiliary power cables included with the RTX 6000 100mm PCIe 6pin (2x3) + PCIe 8pin (2x4): Server support: SR670 Option: 4Z57A26300* SBB: SBB7A16972 Base: SC17A10850 FRU: 01PG437

Note: This cable is used in conjunction with either SBB7A10375 or SBB7A10374.

Auxiliary power cables supplied with the SR670 (configure-to-order) 900mm SR670 Cage 1 power cable Feature: B3Y3 SBB: SBB7A10375 Base: SC17A10876 FRU: 01PG448

Notes:

Feature & SBB also include PCIe data cable Cable also supplies power to the riser

1100mm SR670 Cage 2 power cable Feature: B3Y2 SBB: SBB7A10374 Base: SC17A10863 FRU: 01PG426

Notes:

Feature & SBB also include PCIe data cable Cable also supplies power to the riser

* This cable is also available as a separate option part number for customers that buy the RTX 6000 directly from NVIDIA and wish to use the GPU in the SR670 server.

ThinkSystem and ThinkAgile GPU Summary 49 ThinkSystem NVIDIA Quadro RTX 5000 GPU Shatter the boundaries of what’s possible with the NVIDIA Quadro RTX 5000, powered by NVIDIA Turing GPU to bring real-time ray tracing and accelerated AI to next-generation workflows. Creative and technical professionals can supercharge demanding design and visualization workloads and make more informed decisions faster than ever before. Introduction Part numbers Technical specifications Server support Operating system support Auxiliary power cables

Introduction The Quadro RTX 5000 is equipped with 3072 CUDA cores, 384 Tensor cores, 48 RT cores and 16GB GDDR6 memory. It can render complex models and scenes with physically accurate shadows, reflections, and refractions to empower users with instant insight.

Figure 12. ThinkSystem NVIDIA Quadro RTX 5000 GPU

Part numbers

Table 55. Ordering information

Part number Feature code Description 4X67A17267 B6CH ThinkSystem NVIDIA Quadro RTX 5000 16GB PCIe Active GPU

ThinkSystem and ThinkAgile GPU Summary 50 Technical specifications The following table lists the specifications of the NVIDIA Quadro RTX 5000 GPU.

Table 56. RTX 5000 specifications

Feature Specification GPU Memory 16 GB GDDR6 Memory Interface 256-bit Memory Bandwidth Up to 448 GB/s NVIDIA CUDA® Cores 3072 NVIDIA Tensor Cores 384 NVIDIA RT Cores 48 System Interface PCI Express 3.0 x16 Max Power Consumption 265 W Thermal Solution Active Form Factor 4.4" H x 10.5" L, Dual Slot, Full Height Display Connectors 4x DP 1.4, 1x USB-C Max Simultaneous Displays 4 direct, 4 DP1.4 Multi-Stream Max DP 1.4 Resolution 4x 4096x2160 @ 120 Hz 4x 5120x2880 @ 60 Hz 2x 7680x4320 @ 60 Hz Graphics APIs DirectX 12.0, Shader Model 5.1, OpenGL 4.6, Vulkan 1.1 Compute APIs CUDA, DirectCompute, OpenCL

Server support The following table lists the ThinkSystem servers that are compatible. The numbers listed in the server columns represent the number of GPUs supported.

Table 57. ThinkSystem server support

Dense/ E 1S Intel 2S Intel AMD 4S Intel Blade ) ) 3 ) 0 ) 3 ) / G 6 ) ) ) ) ) ) ) ) ) ) ) ) ) ) V ) ) 2 2 Z X 8 3 2 8 4 3 9 2 6 9 9 0 X 1 2 ) 6 0 3 7 2 3 1 D / 5 0 0 0 9 0 0 9 1 7 1 0 / 0 4 1 / / D 2 D 1 9 5 Y X X Y X X X 7 Y Z X X 2 / D Y X 7 7 / 3 5 7 7 7 7 7 7 7 3 7 7 7 7 1 7 Y 7 7 / F ) / / / / / / / / / / / / / ) ) ) ) / / / D Z 7 2 Y W 4 1 7 3 2 8 1 5 6 8 0 8 9 1 1 8 6 5 6 / 5 9 7 7 2 2 5 5 0 0 0 9 0 0 3 9 0 1 D 6 1 2 5 1 1 ( ( 4 8 4 0

7 4 Y Y X X Y X X X Y Y Y D D X X X X X X X Z Y X ( 2 2

7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 Y 7 7 ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( V V ( ( ( ( ( ( ( P

7

( 0 0 0 0 0 0 0 0 0 5 5 5 5 0 0 0 0 0 0 0 0 0 0 0 0 0 5 5 3 5 7 9 3 5 7 3 5 4 6 5 5 6 5 5 6 3 5 5 5 5 0 5 5 1 2 5 5 5 5 6 6 6 6 6 6 6 8 8 8 9 8 8 5 6 5 8 3 5 2 5 E T T R R T R R R R R R R R R R R R R R R R R D D N N

Description & Part number S S S S S S S S S S S S S S S S S S S S S S S S S S S NVIDIA Quadro RTX 5000, N N N N N N N N N N N 2 N N N N N N N N N N N N N N N 4X67A17267

ThinkSystem and ThinkAgile GPU Summary 51 Operating system support The following table lists the supported operating systems.

Tip: This table is automatically generated based on data from Lenovo ServerProven.

Table 58. Operating system support for ThinkSystem NVIDIA Quadro RTX 5000 16GB PCIe Active GPU, 4X67A17267 ) ) 2 1

n n e e G G ( (

0 0 5 5 6 6 R R

Operating systems S S Microsoft Windows Server 2012 R2 N Y Microsoft Windows Server 2016 Y Y Microsoft Windows Server 2019 Y Y Red Hat Enterprise Linux 7.4 N Y Red Hat Enterprise Linux 7.5 N Y Red Hat Enterprise Linux 7.6 Y Y Red Hat Enterprise Linux 7.7 Y Y Red Hat Enterprise Linux 7.8 Y Y Red Hat Enterprise Linux 8.0 Y Y Red Hat Enterprise Linux 8.1 Y Y Red Hat Enterprise Linux 8.2 Y Y SUSE Linux Enterprise Server 12 SP2 N Y SUSE Linux Enterprise Server 12 SP2 with Xen N Y SUSE Linux Enterprise Server 12 SP3 N Y SUSE Linux Enterprise Server 12 SP3 with Xen N Y SUSE Linux Enterprise Server 12 SP4 Y Y SUSE Linux Enterprise Server 12 SP4 with Xen Y Y SUSE Linux Enterprise Server 15 Y Y SUSE Linux Enterprise Server 15 SP1 Y Y SUSE Linux Enterprise Server 15 SP1 with Xen Y Y SUSE Linux Enterprise Server 15 with Xen Y Y

ThinkSystem and ThinkAgile GPU Summary 52 Auxiliary power cables The RTX 5000 option part number ships with one auxiliary power cable. This cable is needed when the GPU is installed in specific servers, as listed in the following table.

Table 59. Auxiliary power cables (click images to show larger versions)

Auxiliary power cables included with the RTX 5000 option part number 300mm PCIe 6pin (2x3) + PCIe 8pin (2x4): Server support: SR650† SBB: SBB7A00293 Base: SC17A02243 FRU: 01KN070

† For customers who buy the GPU directly from NVIDIA, this cable is also included in the ThinkSystem SR650 GPU Cable Kit, 4XH7A08794

ThinkSystem and ThinkAgile GPU Summary 53 ThinkSystem NVIDIA Quadro P6000 GPU ThinkSystem NVIDIA Quadro P6000 GPU cards are certified with a broad range of sophisticated professional applications, tested by leading workstation manufacturers, and backed by a global team of support specialists. This gives you the peace of mind to focus on doing your best work. Whether you’re developing revolutionary products or telling spectacularly vivid visual stories, Quadro gives you the performance to do it brilliantly. Introduction Part numbers Technical specifications Server support Operating system support Auxiliary power cables

Introduction ThinkSystem NVIDIA Quadro P6000 is the world’s most advanced professional graphics solution ever created, combining the latest GPU, memory and display technologies that result in unprecedented performance and breakthrough capabilities. Professionals across of range of industries can now create their most complex designs, solve the most challenging visualization problems and experience their creations within the most detailed, life-like VR environments.

Figure 13. ThinkSystem NVIDIA Quadro P6000 GPU

Part numbers

Table 60. Ordering information

Part number Feature code Description 7C57A02895 AVRK, B2V5* ThinkSystem NVIDIA Quadro P6000 * Feature code varies by system

ThinkSystem and ThinkAgile GPU Summary 54 Technical specifications The following table lists the specifications of the NVIDIA Quadro P6000 GPU.

Table 61. P6000 specifications

Feature Specification GPU Memory 24 GB GDDR5X Memory Interface 384-bit Memory Bandwidth Up to 432 GB/s NVIDIA CUDA® Cores 3840 System Interface PCI Express 3.0 x16 Max Power Consumption 250 W Thermal Solution Active Form Factor 4.4’’H x 10.5’’ L, Dual Slot, Full Height Display Connectors 4x DP 1.4 + DVI-D DL Max Simultaneous Displays 4 direct, 4 DP1.4 Multi-Stream Max DP 1.4 Resolution 7680 x 4320 @ 30 Hz Max DVI-D DL Resolution 2560 x 1600 @ 60 Hz Graphics APIs Shader Model 5.1, OpenGL 4.54, DirectX 12.05, Vulkan 1.04 Compute APIs CUDA, DirectCompute, OpenCL

Server support The following table lists the ThinkSystem servers that are compatible. The numbers listed in the server columns represent the number of GPUs supported.

Table 62. ThinkSystem server support

Dense/ E 1S Intel 2S Intel AMD 4S Intel Blade ) ) 3 ) 0 ) 3 ) / G 6 ) ) ) ) ) ) ) ) ) ) ) ) ) ) V ) ) 2 2 X Z 8 3 2 8 4 3 9 2 6 9 9 0 X 1 2 6 0 ) 3 2 7 3 1 D / 5 0 0 0 9 0 0 9 1 7 1 0 / 4 1 / / 0 D 2 D 1 9 Y X X Y X X X 7 Y Z X X 2 5 / Y X D 7 7 / 3 5 7 7 7 7 7 7 7 3 7 7 7 7 1 7 7 7 Y / F ) / / / / / / / / / / / / / ) ) ) ) / / / D Z 7 2 Y W 4 1 7 3 2 8 1 5 6 8 0 8 9 1 1 8 6 5 5 9 6 / 7 7 2 2 5 5 0 0 0 9 0 0 3 9 0 1 6 1 2 5 1 1 D ( ( 4 0 4 8

7 4 Y Y X X Y X X X Y Y Y D D X X X X X X X Z Y X ( 2 2

7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 Y 7 7 ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( V V ( ( ( ( ( ( ( P

7

( 0 0 0 0 0 0 0 0 0 5 5 5 5 0 0 0 0 0 0 0 0 0 0 0 0 0 5 5 3 5 7 9 3 5 7 3 5 4 6 5 5 6 5 5 6 3 5 5 5 5 0 5 5 1 2 5 5 5 5 6 6 6 6 6 6 6 8 8 8 9 8 8 5 6 5 8 3 5 2 5 E T T R R T R R R R R R R R R R R R R R R R R D D N N

Description & Part number S S S S S S S S S S S S S S S S S S S S S S S S S S S NVIDIA Quadro P6000, N N N N N 2 N N N N N 2 N N N N N N N N N N N N N N N 7C57A02895

ThinkSystem and ThinkAgile GPU Summary 55 Operating system support The following table lists the supported operating systems.

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

Table 63. Operating system support for ThinkSystem NVIDIA Quadro P6000 24GB PCIe Active GPU, 7C57A02895 ) ) ) ) 2 1 2 1

n n n n e e e e G G G G ( ( ( (

0 0 0 0 5 5 5 5 6 6 5 5 R T R T

Operating systems S S S S Microsoft Windows Server 2012 R2 N N Y Y Microsoft Windows Server 2016 Y Y Y Y Microsoft Windows Server 2019 Y Y Y Y Red Hat Enterprise Linux 6.10 N N Y Y Red Hat Enterprise Linux 6.9 N N Y Y Red Hat Enterprise Linux 7.3 N N Y Y Red Hat Enterprise Linux 7.4 N N Y Y Red Hat Enterprise Linux 7.5 N N Y Y Red Hat Enterprise Linux 7.6 Y Y Y Y Red Hat Enterprise Linux 7.7 Y Y Y Y Red Hat Enterprise Linux 7.8 Y Y Y Y Red Hat Enterprise Linux 8.0 Y Y Y Y Red Hat Enterprise Linux 8.1 Y Y Y Y Red Hat Enterprise Linux 8.2 Y Y Y Y SUSE Linux Enterprise Server 11 SP4 N N Y Y SUSE Linux Enterprise Server 11 SP4 with Xen N N Y Y SUSE Linux Enterprise Server 12 SP2 N N Y Y SUSE Linux Enterprise Server 12 SP2 with Xen N N Y Y SUSE Linux Enterprise Server 12 SP3 N N Y Y SUSE Linux Enterprise Server 12 SP3 with Xen N N Y Y SUSE Linux Enterprise Server 12 SP4 Y Y Y Y SUSE Linux Enterprise Server 12 SP4 with Xen Y Y Y Y SUSE Linux Enterprise Server 15 Y Y Y Y SUSE Linux Enterprise Server 15 SP1 Y Y Y Y SUSE Linux Enterprise Server 15 SP1 with Xen Y Y Y Y SUSE Linux Enterprise Server 15 with Xen Y Y Y Y

ThinkSystem and ThinkAgile GPU Summary 56 Auxiliary power cables The P6000 option part number ships with two auxiliary power cables. These cables are needed when the GPU is installed in specific servers, as listed in the following table.

Table 64. Auxiliary power cables (click images to show larger versions)

Auxiliary power cables included with the GPU option part number 300mm PCIe 6pin (2x3) + PCIe 8pin (2x4): Server support: SR650† SBB: SBB7A00293 Base: SC17A02243 FRU: 01KN070

815mm PCIe 8pin (2x4): Server support: ST550 SBB: SBB7A00295 Base: SC17A12116 FRU: 01KN095

† For customers who buy the GPU directly from NVIDIA, this cable is also included in the ThinkSystem SR650 GPU Cable Kit, 4XH7A08794

ThinkSystem and ThinkAgile GPU Summary 57 ThinkSystem NVIDIA Quadro RTX 4000 GPU Meet the challenge of today’s demanding professional workflows with NVIDIA Quadro RTX 4000, powered by NVIDIA Turing architecture and the NVIDIA RTX platform. The NVIDIA Quadro RTX 4000 delivers GPU accelerated ray tracing, deep learning, and advanced shading in an accessible single slot form factor. It gives designers the power to accelerate their creative efforts with faster time to insight and faster time to solution. Introduction Part numbers Technical specifications Server support Operating system support Auxiliary power cables

Introduction The Quadro RTX 4000 is equipped with 2304 CUDA cores, 288 Tensor Cores, 36 RT cores and 8 GB GDDR6 memory. The single-slot GPU is designed to manage the most intensive AEC, DCC, AI, VR and graphics workloads.

Figure 14. ThinkSystem NVIDIA Quadro RTX 4000 GPU

Part numbers

Table 65. Ordering information

Part number Feature code Description 4X67A14934 B6CG ThinkSystem NVIDIA Quadro RTX 4000 8GB PCIe Active GPU

ThinkSystem and ThinkAgile GPU Summary 58 Technical specifications The following table lists the specifications of the NVIDIA Quadro RTX 4000 GPU.

Table 66. RTX 4000 specifications

Feature Specification GPU Memory 8 GB GDDR6 Memory Interface 256-bit Memory Bandwidth Up to 416 GB/s NVIDIA CUDA® Cores 2304 NVIDIA Tensor Cores 288 NVIDIA RT Cores 36 System Interface PCI Express 3.0 x16 Max Power Consumption 160 W Thermal Solution Active Form Factor 4.4" H x 9.5" L, Single Slot, Full Height Display Connectors 3x DP 1.4, 1x USB-C Max Simultaneous Displays 4 direct, 4 DP1.4 Multi-Stream Max DP 1.4 Resolution 4x 3840x2160 @ 120 Hz 4x 5120x2880 @ 60 Hz 2x 7680x4320 @ 60 Hz Graphics APIs DirectX 12.0, Shader Model 5.1, OpenGL 4.6, Vulkan 1.1 Compute APIs CUDA, DirectCompute, OpenCL

Server support The following table lists the ThinkSystem servers that are compatible. The numbers listed in the server columns represent the number of GPUs supported.

Table 67. ThinkSystem server support

Dense/ E 1S Intel 2S Intel AMD 4S Intel Blade ) ) 3 ) 0 ) 3 ) / G 6 ) ) ) ) ) ) ) ) ) ) ) ) ) ) V ) ) 2 2 Z X 8 3 2 8 4 3 9 2 6 9 9 0 X 1 2 ) 6 0 3 7 2 3 1 D / 5 0 0 0 9 0 0 9 1 7 1 0 / 0 4 1 / / D 2 D 1 9 5 Y X X Y X X X 7 Y Z X X 2 / D Y X 7 7 / 3 5 7 7 7 7 7 7 7 3 7 7 7 7 1 7 Y 7 7 / F ) / / / / / / / / / / / / / ) ) ) ) / / / D Z 7 2 Y W 4 1 7 3 2 8 1 5 6 8 0 8 9 1 1 8 6 5 6 / 5 9 7 7 2 2 5 5 0 0 0 9 0 0 3 9 0 1 D 6 1 2 5 1 1 ( ( 4 8 4 0

7 4 Y Y X X Y X X X Y Y Y D D X X X X X X X Z Y X ( 2 2

7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 Y 7 7 ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( V V ( ( ( ( ( ( ( P

7

( 0 0 0 0 0 0 0 0 0 5 5 5 5 0 0 0 0 0 0 0 0 0 0 0 0 0 5 5 3 5 7 9 3 5 7 3 5 4 6 5 5 6 5 5 6 3 5 5 5 5 0 5 5 1 2 5 5 5 5 6 6 6 6 6 6 6 8 8 8 9 8 8 5 6 5 8 3 5 2 5 E T T R R T R R R R R R R R R R R R R R R R R D D N N

Description & Part number S S S S S S S S S S S S S S S S S S S S S S S S S S S NVIDIA Quadro RTX 4000, N N N N N 2 N N N N 1 3 N N N N N N N N N N N N N N N 4X67A14934

ThinkSystem and ThinkAgile GPU Summary 59 Operating system support The following table lists the supported operating systems.

Tip: This table is automatically generated based on data from Lenovo ServerProven.

Table 68. Operating system support for ThinkSystem NVIDIA Quadro RTX 4000 8GB PCIe Active GPU, 4X67A14934 ) ) ) ) ) ) 2 2 1 1 2 1

n n n n n n e e e e e e G G G G G G ( ( ( ( ( (

0 0 0 0 0 0 3 5 3 5 5 5 6 6 6 6 5 5 R R T R R T

Operating systems S S S S S S Microsoft Windows Server 2012 R2 N N N Y Y Y Microsoft Windows Server 2016 Y Y Y Y Y Y Microsoft Windows Server 2019 Y Y Y Y Y Y Red Hat Enterprise Linux 7.4 N N N Y Y Y Red Hat Enterprise Linux 7.5 N N N Y Y Y Red Hat Enterprise Linux 7.6 Y Y Y Y Y Y Red Hat Enterprise Linux 7.7 Y Y Y Y Y Y Red Hat Enterprise Linux 7.8 Y Y Y Y Y Y Red Hat Enterprise Linux 8.0 Y Y Y Y Y Y Red Hat Enterprise Linux 8.1 Y Y Y Y Y Y Red Hat Enterprise Linux 8.2 Y Y Y Y Y Y SUSE Linux Enterprise Server 12 SP2 N N N Y Y Y SUSE Linux Enterprise Server 12 SP2 with Xen N N N Y Y Y SUSE Linux Enterprise Server 12 SP3 N N N Y Y Y SUSE Linux Enterprise Server 12 SP3 with Xen N N N Y Y Y SUSE Linux Enterprise Server 12 SP4 Y Y Y Y Y Y SUSE Linux Enterprise Server 12 SP4 with Xen Y Y Y Y Y Y SUSE Linux Enterprise Server 15 Y Y Y Y Y Y SUSE Linux Enterprise Server 15 SP1 Y Y Y Y Y Y SUSE Linux Enterprise Server 15 SP1 with Xen Y Y Y Y Y Y SUSE Linux Enterprise Server 15 with Xen Y Y Y Y Y Y

ThinkSystem and ThinkAgile GPU Summary 60 Auxiliary power cables The RTX 4000 option part number ships with three auxiliary power cables. These cables are needed when the GPU is installed in specific servers, as listed in the following table.

Table 69. Auxiliary power cables (click images to show larger versions)

Auxiliary power cables included with the GPU option part number 300mm PCIe 8pin (2x4): Server support: SR630 SBB: SBB7A14675 Base: SC17A42754 FRU: 02JG463

300mm 2x 8pin (2x4) cable Server support: SR650† SBB: SBB7A05971 Base: SC17A30059 FRU: 01KR723

815mm PCIe 8pin (2x4): Server support: ST550 SBB: SBB7A00295 Base: SC17A12116 FRU: 01KN095

† For customers who buy the GPU directly from NVIDIA, this cable is also included in the ThinkSystem SR650 GPU Cable Kit, 4XH7A08794

ThinkSystem and ThinkAgile GPU Summary 61 ThinkSystem NVIDIA Quadro P2200 GPU ThinkSystem Quadro P2200 is the perfect balance of performance, compelling features, and compact form factor delivering incredible creative experience and productivity across a variety of professional 3D applications. The P2200 is a follow-on to the successful P2000 GPU. Introduction Part numbers Technical specifications Server support Operating system support Auxiliary power cables

Introduction ThinkSystem Quadro P2200 features an Pascal GPU with 1280 CUDA cores, large 5 GB GDDR5X on- board memory, and the power to drive up to four 5K (5120x2880 @ 60Hz) displays natively. These features make is an excellent choice to help accelerate product development and content creation workflows with large scenes and models.

Figure 15. ThinkSystem NVIDIA Quadro P2200 GPU

Part numbers

Table 70. Ordering information

Part number Feature code Description 4X67A14935 B7JW ThinkSystem NVIDIA Quadro P2200 5GB PCIe Active GPU

ThinkSystem and ThinkAgile GPU Summary 62 Technical specifications The following table lists the specifications of the NVIDIA Quadro P2200 GPU.

Table 71. P2200 specifications

Feature Specification GPU Memory 5 GB GDDR5X Memory Interface 160-bit Memory Bandwidth Up to 200 GB/s NVIDIA CUDA® Cores 1280 System Interface PCI Express 3.0 x16 Max Power Consumption 75 W Thermal Solution Active Form Factor 4.4”H x 7.9” L, Single Slot Display Connectors 4x DP 1.4 Max Simultaneous Displays 4 direct, 4 DP1.4 Multi-Stream Display Resolution 4x 4096x2160 @ 120Hz 4x 5120x2880 @ 60Hz Graphics APIs Shader Model 5.1, OpenGL 4.6, DirectX 12.0, Vulkan 1.1 Compute APIs CUDA, DirectCompute, OpenCL

Server support The following table lists the ThinkSystem servers that are compatible. The numbers listed in the server columns represent the number of GPUs supported.

Table 72. ThinkSystem server support

Dense/ E 1S Intel 2S Intel AMD 4S Intel Blade ) ) 3 ) 0 3 ) / ) G 6 ) ) ) ) ) ) ) ) ) ) ) ) ) ) V ) ) 2 2 X Z 8 3 2 8 4 3 9 2 6 9 9 0 X 1 2 ) 6 0 3 2 7 3 1 D / 5 0 0 0 9 0 0 9 1 7 1 / 0 0 4 1 / / D 2 1 D 9 5 Y X X Y X X X 7 Y Z X X 2 / D Y X 7 3 7 / 5 7 7 7 7 7 7 7 3 7 7 7 7 1 Y F 7 7 7 / ) / / / / / / / / / / / / / ) ) ) ) / / / D Z 7 2 Y W 4 1 7 3 2 8 1 5 6 8 0 8 9 1 1 8 6 5 / 6 5 9 7 7 2 2 D 5 5 0 0 0 9 0 0 3 9 0 1 6 1 2 5 1 1 ( ( 8 4 4 0

7 4 Y Y X X Y X X X Y Y Y D D X X X X X X X Z Y X ( 2 2

7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 Y 7 7 ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( V V ( ( ( ( ( ( ( P

7

( 0 0 0 0 0 0 0 0 0 5 5 5 5 0 0 0 0 0 0 0 0 0 0 0 0 0 5 5 3 5 7 9 3 5 7 3 5 4 6 5 5 6 5 5 6 3 5 5 5 5 0 5 5 1 2 5 5 5 5 6 6 6 6 6 6 6 8 8 8 9 8 8 5 6 5 8 3 5 2 5 E T T R R T R R R R R R R R R R R R R R R R R D D N N

Description & Part number S S S S S S S S S S S S S S S S S S S S S S S S S S S NVIDIA Quadro P2200, N N 1 N N 2 N N N N 1 N N N N 1 N N N N N N N N N N N 4X67A14935

ThinkSystem and ThinkAgile GPU Summary 63 Operating system support The following table lists the supported operating systems.

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

Table 73. Operating system support for ThinkSystem NVIDIA Quadro P2200 5GB PCIe Active GPU, 4X67A14935 ) ) ) ) 2 1 2 1

n n n n e e e e G G G G ( ( ( (

5 0 0 0 0 0 4 3 3 5 5 5 6 6 6 2 5 5 T R R T R T

Operating systems S S S S S S Microsoft Windows Server 2012 R2 N N N N Y Y Microsoft Windows Server 2016 Y Y Y Y Y Y Microsoft Windows Server 2019 Y Y Y Y Y Y Red Hat Enterprise Linux 7.5 Y N N N Y Y Red Hat Enterprise Linux 7.6 N N Y Y Y Y Red Hat Enterprise Linux 7.7 Y N Y Y Y Y Red Hat Enterprise Linux 7.8 Y N Y Y Y Y Red Hat Enterprise Linux 8.0 Y N Y Y Y Y Red Hat Enterprise Linux 8.1 Y N Y Y Y Y Red Hat Enterprise Linux 8.2 Y N Y Y Y Y SUSE Linux Enterprise Server 12 SP2 N N N N Y Y SUSE Linux Enterprise Server 12 SP2 with Xen N N N N Y Y SUSE Linux Enterprise Server 12 SP3 Y N N N Y Y SUSE Linux Enterprise Server 12 SP3 with Xen Y N N N Y Y SUSE Linux Enterprise Server 12 SP4 N N Y Y Y Y SUSE Linux Enterprise Server 12 SP4 with Xen N N Y Y Y Y SUSE Linux Enterprise Server 12 SP5 Y N N N N N SUSE Linux Enterprise Server 12 SP5 with Xen Y N N N N N SUSE Linux Enterprise Server 15 Y N Y Y Y Y SUSE Linux Enterprise Server 15 SP1 Y Y Y Y Y Y SUSE Linux Enterprise Server 15 SP1 with Xen Y Y Y Y Y Y SUSE Linux Enterprise Server 15 with Xen Y N Y Y Y Y

Auxiliary power cables The P2200 does not require an auxiliary power cable.

ThinkSystem and ThinkAgile GPU Summary 64 ThinkSystem NVIDIA Quadro P620 GPU The NVIDIA Quadro P620 is certified with a broad range of sophisticated professional applications, tested by leading workstation manufacturers, and backed by a global team of support specialists. Introduction Part numbers Technical specifications Server support Operating system support Auxiliary power cables

Introduction The NVIDIA Quadro P620 workstation graphics board is powered by NVIDIA Pascal GPU technology and 2 GB of GDDR5 memory. The Quadro P620 graphics board is targeted for professional CAD, DCC and visualization designers, engineers and users. Get the budget friendly highest performing graphics board to drive today’s demanding professional workflows in a compact footprint. Flexible single-slot and low-profile form factor makes this card compatible with even the most space and power constrained workstation chassis.

Figure 16. ThinkSystem NVIDIA Quadro P620 GPU

Part numbers

Table 74. Ordering information

Part number Feature code Description 4X67A11584 B31D ThinkSystem NVIDIA Quadro P620

ThinkSystem and ThinkAgile GPU Summary 65 Technical specifications The following table lists the specifications of the NVIDIA Quadro P620 GPU.

Table 75. Technical specifications

Feature Specifications GPU Memory 2 GB GDDR5 Memory Interface 128-bit Memory Bandwidth Up to 80 GB/s NVIDIA CUDA Cores 512 System Interface PCI Express 3.0 x16 Max Power Consumption 40 W Thermal Solution Active Display Connectors 4x mDP 1.4 Max Simultaneous Displays 4 direct, 4 DP 1.4 Multi-Stream Display Resolution 4x 4096x2160 @ 60Hz 4x 5120x2880 @ 60Hz Graphics APIs Shader Model 5.1, OpenGL 4.5, DirectX 12.0, Vulkan 1.0 Compute APIs CUDA, DirectCompute, OpenCL

Server support The following table lists the ThinkSystem servers that are compatible. The numbers listed in the server columns represent the number of GPUs supported.

Table 76. ThinkSystem server support

Dense/ E 1S Intel 2S Intel AMD 4S Intel Blade ) ) 3 ) 0 ) 3 ) / G 6 ) ) ) ) ) ) ) ) ) ) ) ) ) ) V ) ) 2 2 X Z 8 3 2 8 4 3 9 2 6 9 9 0 X 1 2 ) 6 0 3 2 7 3 1 D / 5 0 0 0 9 0 0 9 1 7 1 0 / 0 4 1 / / D 2 D 1 9 5 Y X X Y X X X 7 Y Z X X 2 / D Y X 7 7 / 3 5 7 7 7 7 7 7 7 3 7 7 7 7 1 7 Y 7 7 / F ) / / / / / / / / / / / / / ) ) ) ) / / / D Z 7 2 Y W 4 1 7 3 2 8 1 5 6 8 0 8 9 1 1 8 6 5 6 / 5 9 7 7 2 2 5 5 0 0 0 9 0 0 3 9 0 1 D 6 1 2 5 1 1 ( ( 4 8 4 0

7 4 Y Y X X Y X X X Y Y Y D D X X X X X X X Z Y X ( 2 2

7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 Y 7 7 ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( V V ( ( ( ( ( ( ( P

7

( 0 0 0 0 0 0 0 0 0 5 5 5 5 0 0 0 0 0 0 0 0 0 0 0 0 0 5 5 3 5 7 9 3 5 7 3 5 4 6 5 5 6 5 5 6 3 5 5 5 5 0 5 5 1 2 5 5 5 5 6 6 6 6 6 6 6 8 8 8 9 8 8 5 6 5 8 3 5 2 5 E T T R R T R R R R R R R R R R R R R R R R R D D N N

Description & Part number S S S S S S S S S S S S S S S S S S S S S S S S S S S NVIDIA Quadro P620, N 1 1 N 1 2 N N N N 3 3 N 3 6 3 8 N N N N N N N N N N 4X67A11584

ThinkSystem and ThinkAgile GPU Summary 66 Operating system support The following table lists the supported operating systems.

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

Table 77. Operating system support for ThinkSystem NVIDIA Quadro P620 PCIe Active GPU, 4X67A11584 ) ) ) ) ) ) 2 2 1 1 2 1

n n n n n n e e e e e e G G G G G G ( ( ( ( ( (

0 5 5 5 5 0 0 0 0 0 0 0 5 3 4 5 6 3 5 3 5 5 0 5 5 2 6 6 6 6 6 6 6 6 2 5 5 5 R T T R R R R R R T R R T

Operating systems S S S S S S S S S S S S S Microsoft Windows Server 2016 Y Y Y Y Y Y Y Y Y Y Y Y Y Microsoft Windows Server 2019 Y Y Y Y Y Y Y Y Y Y Y Y Y Red Hat Enterprise Linux 6.10 N N N N N N N N N N Y Y Y Red Hat Enterprise Linux 6.9 N N N N N N N N N N Y Y Y Red Hat Enterprise Linux 7.3 N N N N N N N N N N Y Y Y Red Hat Enterprise Linux 7.4 N N N N N N N N N N Y Y Y Red Hat Enterprise Linux 7.5 Y Y Y N N N N N N N Y Y Y Red Hat Enterprise Linux 7.6 N N N Y N Y N Y Y Y Y Y Y Red Hat Enterprise Linux 7.7 Y Y Y Y N Y N Y Y Y Y Y Y Red Hat Enterprise Linux 7.8 Y Y Y Y N Y N Y Y Y Y Y Y Red Hat Enterprise Linux 7.9 N N Y N N N N N N N N N N Red Hat Enterprise Linux 8.0 Y Y Y Y N Y N Y Y Y Y Y Y Red Hat Enterprise Linux 8.1 Y Y Y Y N Y N N N N N N N Red Hat Enterprise Linux 8.2 Y Y Y Y N Y N Y Y Y Y Y Y SUSE Linux Enterprise Server 11 SP4 N N N N N N N N N N Y Y Y SUSE Linux Enterprise Server 11 SP4 with Xen N N N N N N N N N N Y Y Y SUSE Linux Enterprise Server 12 SP2 N N N N N N N N N N Y Y Y SUSE Linux Enterprise Server 12 SP2 with Xen N N N N N N N N N N Y Y Y SUSE Linux Enterprise Server 12 SP3 Y Y N N N N N N N N Y Y Y SUSE Linux Enterprise Server 12 SP3 with Xen Y Y N N N N N N N N Y Y Y SUSE Linux Enterprise Server 12 SP4 N N N Y N Y N Y Y Y Y Y Y SUSE Linux Enterprise Server 12 SP4 with Xen N N N Y N Y N Y Y Y Y Y Y SUSE Linux Enterprise Server 12 SP5 Y N N Y N N N N N N N N N SUSE Linux Enterprise Server 12 SP5 with Xen Y N N Y N N N N N N N N N SUSE Linux Enterprise Server 15 Y Y N N N N N Y Y Y Y Y Y SUSE Linux Enterprise Server 15 SP1 Y Y Y Y Y Y Y Y Y Y Y Y Y SUSE Linux Enterprise Server 15 SP1 with Xen Y Y Y Y Y Y Y Y Y Y Y Y Y SUSE Linux Enterprise Server 15 SP2 N N N Y N N N N N N N N N SUSE Linux Enterprise Server 15 SP2 with Xen N N N Y N N N N N N N N N SUSE Linux Enterprise Server 15 with Xen Y Y N N N N N Y Y Y Y Y Y

ThinkSystem and ThinkAgile GPU Summary 67 Auxiliary power cables The P620 does not require an auxiliary power cable.

Related product families Product families related to this document are the following: GPU adapters

ThinkSystem and ThinkAgile GPU Summary 68 Notices

Lenovo may not offer the products, services, or features discussed in this document in all countries. Consult your local Lenovo representative for information on the products and services currently available in your area. Any reference to a Lenovo product, program, or service is not intended to state or imply that only that Lenovo product, program, or service may be used. Any functionally equivalent product, program, or service that does not infringe any Lenovo intellectual property right may be used instead. However, it is the user's responsibility to evaluate and verify the operation of any other product, program, or service. Lenovo may have patents or pending patent applications covering subject matter described in this document. The furnishing of this document does not give you any license to these patents. You can send license inquiries, in writing, to:

Lenovo (United States), Inc. 1009 Think Place - Building One Morrisville, NC 27560 U.S.A. Attention: Lenovo Director of Licensing

LENOVO PROVIDES THIS PUBLICATION ”AS IS” WITHOUT WARRANTY OF ANY KIND, EITHER EXPRESS OR IMPLIED, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF NON-INFRINGEMENT, MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE. Some jurisdictions do not allow disclaimer of express or implied warranties in certain transactions, therefore, this statement may not apply to you.

This information could include technical inaccuracies or typographical errors. Changes are periodically made to the information herein; these changes will be incorporated in new editions of the publication. Lenovo may make improvements and/or changes in the product(s) and/or the program(s) described in this publication at any time without notice.

The products described in this document are not intended for use in implantation or other life support applications where malfunction may result in injury or death to persons. The information contained in this document does not affect or change Lenovo product specifications or warranties. Nothing in this document shall operate as an express or implied license or indemnity under the intellectual property rights of Lenovo or third parties. All information contained in this document was obtained in specific environments and is presented as an illustration. The result obtained in other operating environments may vary. Lenovo may use or distribute any of the information you supply in any way it believes appropriate without incurring any obligation to you.

Any references in this publication to non-Lenovo Web sites are provided for convenience only and do not in any manner serve as an endorsement of those Web sites. The materials at those Web sites are not part of the materials for this Lenovo product, and use of those Web sites is at your own risk. Any performance data contained herein was determined in a controlled environment. Therefore, the result obtained in other operating environments may vary significantly. Some measurements may have been made on development-level systems and there is no guarantee that these measurements will be the same on generally available systems. Furthermore, some measurements may have been estimated through extrapolation. Actual results may vary. Users of this document should verify the applicable data for their specific environment.

© Copyright Lenovo 2021. All rights reserved.

This document, LP0768, was created or updated on February 21, 2021. Send us your comments in one of the following ways: Use the online Contact us review form found at: http://lenovopress.com/LP0768 Send your comments in an e-mail to: [email protected] This document is available online at http://lenovopress.com/LP0768.

ThinkSystem and ThinkAgile GPU Summary 69 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: 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. DirectX®, 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.

ThinkSystem and ThinkAgile GPU Summary 70