Lenovo NVIDIA HGX Rubin NVL8

The Lenovo NVIDIA HGX Rubin NVL8 is a rack solution built on the NVIDIA HGX™ Rubin NVL8 modular architecture, which delivers groundbreaking advancements in accelerating computing, leading the new era of AI with optimized compute and increased memory.

The NVIDIA HGX Rubin NVL8 is the NVIDIA rack-scale AI offering. It is an end-to-end NVIDIA reference design adopted by the market.

The NVIDIA HGX Rubin NVL8 is a platform designed for AI reasoning performance and efficiency, a scalable building block approach design, starting with a single HGX Rubin NVL8 system and scaling to as many as nine systems in a 48U NVIDIA MGX™ rack.

The HGX™ Rubin NVL8 uses a fully direct-water-cooled, zero-airflow system design to manage heat from its processors, GPUs, NVLink Switches, and networking components. Support for inlet-water temperatures up to 45°C enables better performance per watt and dense GPU computing within a compact 2U chassis, helping data centers deploy high-performance AI infrastructure while reducing dependence on system-level air cooling.

Figure 1. Lenovo NVIDIA HGX Rubin NVL8

Lenovo NVIDIA HGX Rubin NVL8 server supported in a 48U NVIDIA MGX rack fully integrates eight NVIDIA Rubin GPUs. These GPUs are interconnected through four NVIDIA® NVLink Switches, leveraging direct-water-cooled cooling architecture to support a tightly connected platform for demanding AI workloads, forming a single high performance compute domain ideal for large‑scale multimodal reasoning, high‑performance inference and LLM training workloads.

The Lenovo NVIDIA HGX Rubin NVL8 is a dense 2U fully liquid-cooled AI solution. Ideal for customers looking for a modular design, node-level growth, and a familiar HGX‑class deployment model they can roll out in smaller steps.

• CPU architecture based on x86 Intel
• Enterprise deployment pattern
• Fully liquid cooled nodes
• Modularity & phased procurement

Turnkey AI Rack Design

Engineered for large-scale Artificial Intelligence (AI) and High Performance Computing (HPC), Lenovo NVIDIA HGX Rubin NVL8 excels in intensive simulations and complex modeling. It is designed to handle modeling, training, simulation, rendering, financial tech, scientific research, technical computing, grid deployments, and analytics workloads in various fields such as research, life sciences, energy, engineering, and financial simulation.

The Lenovo NVIDIA HGX Rubin NVL8 uses a fully direct-water-cooled, zero-airflow system design to manage heat from its processors, GPUs, NVLink Switches, and networking components. Support for inlet-water temperatures up to 45°C enables better performance per watt and dense GPU computing within a compact 2U chassis, helping data centers deploy high-performance AI infrastructure while reducing dependence on system level air cooling.

Eight NVIDIA Rubin GPUs are interconnected through four NVIDIA NVLink Switches, creating a tightly connected platform. Eight NVIDIA ConnectX-9 SuperNICs and NVIDIA BlueField4 DPU provide the high-speed networking infrastructure needed to support AI training, inference, and other data-intensive workloads.

Direct water-cooled solutions are factory-integrated and are re-tested at the rack-level to ensure that a rack can be directly deployed at the customer site. This careful and consistent quality testing has been developed as a result of over a decade of experience designing and deploying DWC solutions to the very highest standards.

Scalability and performance

The Lenovo NVIDIA HGX Rubin NVL8 is based on Intel Xeon 6 processors and NVIDIA Rubin GPU design which provides the following features for high-performance, AI computing:

• The Lenovo NVIDIA HGX Rubin NVL8 rack server
  • Two Intel Xeon 6 processors, 64 cores, 350W
  • 32x DDR5 6400MHz up to 4TB
  • 2x M.2 boot device, hot swap
  • Up to 8x E1.S drives, for internal storage
  • Eight NVIDIA Rubin GPUs are interconnected through four NVIDIA NVLink Switches
    • GPU Bay with HGX Rubin NVL8 baseboard
      • 8x CX9 IO Switches w/ 8x OSFP ports
      • 8x Rubin SXM8 up to 2300W
      • 4x NVSwitches w/ CX9 fabric management
  • Eight NVIDIA ConnectX-9 SuperNICs and NVIDIA BlueField-4 DPU provide the high-speed networking
    • 1x BF-4 B4240L FH3/4L DPU DC-SCM with AMI OpenBMC
  • Direct-water-cooled, zero-airflow system to manage heat from its processors, GPUs, NVLink Switches, and networking components
  • 24 kW System power
  • A 1GbE RJ45 management interface provides BMC management connectivity
• Lenovo 48U MGX Rack Solution with Lenovo NVIDIA HGX Rubin NVL8
  • Max 9x Lenovo NVIDIA HGX Rubin NVL8 rack servers
  • 4x 3RU Power shelf, 110kW 
    • 3 phase PSU
    • 415VAC-480VAC input
    • Air cooled
    • 4 power shelves for 4M3 redundancy
    • Dual 100 A feeds per shelf
  • LC busbar 5000A
  • New manifolds to support Universal Quick Disconnects (UQDs)
  • Rack Sustained power design ~225kW
• The rack includes one NVIDIA SN2201 Top-of-Rack management switches to provide the rack management network.

Each fully populated rack delivers 72 NVIDIA Rubin GPUs across nine Rubin NVL8 nodes, with each 8-GPU node interconnected via sixth-generation NVIDIA NVLink. The modular architecture provides high-performance compute for large-scale AI training, reasoning, and inference workloads.

Energy efficiency

The direct water cooled solution offers the following energy efficiency features to save energy, reduce operational costs, increase energy availability, and contribute to a green environment:

• Water cooling eliminates power that is drawn by cooling fans in the enclosure and dramatically reduces the required air movement in the server room, which also saves power in the data center due to the reduced need for air conditioning.
• Zero airflow design, 100% direct water cooled
• Up to 45ºC inlet water, field serviceable
• Up to 4x 110kW 100A Air‑Cooled Power Shelf with 4M3 redundancy
• The processors and other microelectronics are run at lower temperatures because they are water cooled, which uses less power, and allows for higher performance.

Manageability and security

The following powerful systems management features simplify local and remote management of the NVIDIA HGX Rubin NVL8 system:

• Our Confluent management system provides an interface designed to abstract the users from the complexity of HPC cluster orchestration and AI workloads management, making open-source HPC software consumable for every customer.
• Integrated Trusted Platform Module (TPM) 2.0 support enables advanced cryptographic functionality, such as digital signatures and remote attestation.
• Supports Secure Boot to ensure only a digitally signed operating system can be used.
• Industry-standard Advanced Encryption Standard (AES) NI support for faster, stronger encryption.
• For management: BMC Server Management chip (ASPEED AST2600) and HMC.
• Power shelves are located at top and bottom of the rack (110kW each).
• The Baseboard Management Controller (BMC) in the rack server tray is used in conjunction with the Host Module Controller (HMC) to provide Out-of-band management.

Availability and serviceability

The Lenovo NVIDIA HGX Rubin NVL8 rack server provides the following features to simplify serviceability and increase system uptime:

• Designed to run 24 hours a day, 7 days a week
• Power Shelves to minimize downtime as current is shared between multiple shelves.
• Toolless cover removal on the nodes provides easy access to upgrades and serviceable parts such as adapters and drives.
• The system is designed to support non-disruptive firmware upgrades.
• FRUs (Field Replaceable Unit) are supported by the design. FRUs are parts that can be field replaceable. Field repair is supported with appropriately defined and stocked FRUs (Field Replaceable Units).
• All cabling connectors will include a latch and a keyed design to prevent users from plugging them in backwards.
• External debug console port. Dedicated OS and BMC networking ports provide continuous access for diagnostics.
• There is a three-year customer replaceable unit and onsite limited warranty, with next business day 9x5 coverage. Optional warranty upgrades and extensions are available.

The table below shows the main system components of the rack scale solution. This is to depict an overall hardware configurability, and capability, for the complex set of feature options that the system needs to be able to support in hardware. Not all configurable combinations, or feature options, are shown here; such as DIMM memory options, TPM, or storage options as examples. Also not shown are the TOR switches which are dependent on customer network topology configuration.

Topics in this section:

• Standard specifications - Lenovo MGX 48U Rack for HGX Rubin NVL8
• Standard specifications - Lenovo NVIDIA HGX Rubin NVL8

Standard specifications - Lenovo MGX 48U Rack for HGX Rubin NVL8

The following table lists the standard specifications of the Lenovo MGX 48U Rack for HGX Rubin NVL8.

The Lenovo NVIDIA HGX Rubin NVL8 rack servers, Power Shelves and Top of Rack switch are all integrated and rack mounted. Below is a high-level rack overview of the fully populate Lenovo MGX 48U Rack for HGX Rubin NVL8 configuration.

Figure 2. Front view of the Lenovo MGX 48U Rack for HGX Rubin NVL8

Standard specifications - Lenovo NVIDIA HGX Rubin NVL8

The following table lists the standard specifications of the Lenovo NVIDIA HGX Rubin NVL8 Rack Server.

Notes:

Onsite spares (Compute trays and NVL Switch Trays) would be needed to support an onsite Next Business Day Service Level target. 2% of the order recommended quantity. For Compute Tray base alternatives available for selection, see the Spare Compute Trays section.

The front of the Lenovo NVIDIA HGX Rubin NVL8 server is shown in the following figure.

Figure 3. Front view of the Lenovo NVIDIA HGX Rubin NVL8

The following figure shows the main components on the rear of the Rack Server configuration that includes Quick Disconnect (QD) ports and Busbar clip.

Figure 4. Rear view of the Lenovo NVIDIA HGX Rubin NVL8

The following figure shows key components internal to the Rack Server.

Figure 5. Inside view of the Lenovo NVIDIA HGX Rubin NVL8

The following figure shows the architectural block diagram of the Lenovo NVIDIA HGX Rubin NVL8 rack server, showing the major components and their connections.

The architecture integrates a dual-socket Intel GNR CPU host platform with an eight-GPU Rubin baseboard. The main components include eight Rubin GPUs connected through NVLink switches, CX9 network interfaces with OSFP connectivity, a PCIe switch board, BF4 DPU, OCP adapter, NVMe storage backplane, DC-SCM management module with BMC, and dedicated system management interfaces. The server also supports eight front E1.s NVMe drives and two M.2 NVMe boot drives. This configuration optimizes network connectivity while maximizing computational efficiency, ensuring cost-effectiveness in performance.

Figure 6. 2U Lenovo NVIDIA HGX Rubin NVL8 system architectural block diagram

For more information about the two configurations, see the Configurations section.

The NVIDIA HGX Rubin NVL8 rack is configured by using the configure-to-order (CTO) process with the Lenovo Cluster Solutions configurator (x-config) or in Lenovo Data Center Solution Configurator (DCSC).

The following table lists the base CTO models and base feature codes

The Lenovo NVIDIA HGX Rubin NVL8 node is equipped with high-performance GPUs. It is RoHS complaint and meets all the environmental certifications.

There are two primary configurations offered with the NVIDIA HGX Rubin NVL8.

• Rack: 1x MGX 48U Rack with Top Manifold feed (Bottom feed option available), 54V DC Busbar
  • VR NVL8/72 Top Feed Manifold L&R, CFXU
• Rack: 1x MGX 48U Rack with Bottom Manifold feed (Top feed option available), 54V DC Busbar
  • NVL8/72 Bottom Feed Manifold L&R, CGVB

Block diagrams for 2U Rack Server leveraged in these configurations are shown in the System architecture section.

For more information about the Lenovo NVIDIA HGX Rubin NVL8, see the Lenovo NVIDIA HGX Rubin NVL HG658X V4 datasheet.

The following table lists the Configurators that can be leveraged to create the above configurations.

To guarantee a consistent next business day response objective and minimize potential downtime, we strongly recommend the purchase of on-site spare rack servers and options. Lenovo recommends a minimum spare-to-production ratio of 2%, however the ideal number of spares depends on the scale of your deployment. Contact Lenovo Services for tailored guidance on the appropriate number of spares.

The following Compute Tray Base alternatives are available for selection:

The following Managed Switch alternatives are available for selection:

The Lenovo NVIDIA HGX Rubin NVL8 includes two Intel Xeon 6776P processors. The processors are standard in all configurations.

Each processor has the following features:

• 64 cores and 128 threads
• Intel Xeon 6 processor, formerly codenamed Granite Rapids
• 3 GHz base frequency
• Up to 3.9 GHz maximum turbo frequency
• 336 MB L3 cache
• 350 W TDP
• Support for DDR5 memory with 8 memory channels
• Support for up to 4 TB of memory per processor, dependent on memory type and system implementation
• 88 PCIe 5.0 lanes
• Support for CXL 2.0
• Four Intel UPI links at up to 24 GT/s for processor-to-processor connectivity
• Intel Hyper-Threading Technology
• Intel Advanced Matrix Extensions (Intel AMX)
• Intel AVX-512 and Intel Deep Learning Boost for accelerated AI and HPC workloads
• Intel VT-x and Intel VT-d virtualization technologies
• FCLGA4710 socket package

The NVIDIA Vera processor is integrated using Ball Grid Array (BGA) technology and is soldered onto the NVIDIA VR NVL72 HPM board. To replace the CPU, it is necessary to replace the entire module.

The NVIDIA Rubin NVL78 HPM boards incorporate processor memory via DDR5x modules.

Features of the memory subsystem are as follows:

• 2TB of DDR5X ECC memory in a fixed configuration (32x 64GB DDR5 6400)

The Single Error Correction (SEC) and Double Error Detection (DED) algorithms as part of the ECC are designed to mitigate multi-bit aliasing and correct errors effectively. Furthermore, features such as parity, scrubbing, poisoning, and page off-lining are implemented to enhance Reliability, Availability, and Serviceability (RAS).

The NVIDIA VR NVL72 HPM boards support four onboard NVIDIA Rubin GPUs.

The Lenovo NVIDIA HGX Rubin NVL8 node supports the following drives.

• Up to 8x E1.S drives for data storage.
  • These are front accessible.
• 2x M.2 NVMe drive for OS boot
  • This is an internal drive - not front accessible and non-hot-swap.

The following image shows drives that come standard in the Compute Tray and their location.

Figure 8. Lenovo NVIDIA HGX Rubin NVL8 internal drive bays

Each of the 8 drive bays support the following:

• E1.S 3.84TB or 7.68TB drives

Configuration notes:

• E1.S installation rule:
• The 2U Server Rack supports E1.S NVMe SSDs for data and M.2 NVMe SSD for OS boot
• The drives are connected to onboard controllers; RAID functionality is provided using the operating system features
• The same PCIe lanes can be used either for PCIe adapters or for NVMe drives.

The feature codes to select the appropriate boot drives are listed in the following table.

The following table lists the drive options for internal storage of the Server.

The NVIDIA HGX Rubin NVL8 server supports up to 8x front-accessible GPU slots and 1x PCIe slot:

• Slots 1: PCIe 6.0 x16 slot connected to CPU 0/ CPU 1

The location of the slots is shown in the following figure.

Figure 9. NVIDIA HGX Rubin NVL8 PCIe slots

Slot availability is based on the configuration selected. Derived parts come standard within the  7DQLCTO1WW Lenovo NVIDIA HGX Rubin NVL8 BOM.

PCIe adapters and data drives share the same PCIe connections to the processors. The following table lists the supported data and boot drives.

For more information, including the supported transceivers and cables, see the adapter product guide:

• ConnectX-9: https://lenovopress.lenovo.com/lp2561-thinksystem-nvidia-connectx-9-9180-800gb-1-port-osfp-adapter

For more information, including the supported transceivers and cables, see the adapter product guide:

• ConnectX-9: https://lenovopress.lenovo.com/lp2561-thinksystem-nvidia-connectx-9-9180-800gb-1-port-osfp-adapter

The Lenovo NVIDIA HGX Rubin NVL8 offering supports the two most common fluids used the de-ionized water and PG25.

The customer can choose the coolant they want to use:

• Deionized (DI) water (recommended); Rack ships without water
• PG25 (glycol-based)

One of the most notable features of the Lenovo NVIDIA HGX Rubin NVL8 offering is direct water cooling. Direct water cooling (DWC) is achieved by circulating the cooling water directly through cold plates that contact the GPUs, CPU and memory in the rack server chassis.

There are several components’ temperatures monitored to properly drive the eight system fan pairs. They are OSFP transceivers, QSFP transceivers, BlueField-3, Power Distribution Board components, ambient air, BMC, HMC, M.2, and E1.S components.

One of the main advantages of direct water cooling is the water can be relatively warm and still be effective because water conducts heat much more effectively than air. Depending on the environmentals like water and air temperature, effectively 90% of the heat can be removed by water cooling; in configurations that stay slightly below that, the rest can be easily managed by a standard computer room air conditioner.

Allowable inlet temperatures for the water can be as high as 45°C (113°F) with the NVIDIA HGX Rubin NVL8 for real-world applications. In most climates, water-side economizers can supply water at temperatures below 45°C for most of the year. This ability allows the data center chilled water system to be bypassed thus saving energy because the chiller is the most significant energy consumer in the data center. Typical economizer systems, such as dry-coolers, use only a fraction of the energy that is required by chillers, which produce 6-10°C (43-50°F) water. The facility energy savings are the largest component of the total energy savings that are realized when the NVIDIA HGX Rubin NVL8 is deployed.

The advantages of the use of water cooling over air cooling result from water’s higher specific heat capacity, density, and thermal conductivity. These features allow water to transmit heat over greater distances with much less volumetric flow and reduced temperature difference as compared to air.

For cooling IT equipment, this heat transfer capability is its primary advantage. Water has a tremendously increased ability to transport heat away from its source to a secondary cooling surface, which allows for large, more optimally designed radiators or heat exchangers rather than small, inefficient fins that are mounted on or near a heat source, such as a CPU.

The NVIDIA HGX Rubin NVL8 offering uses the benefits of water by distributing it directly to the highest heat generating node subsystem components. That energy savings results from the removal of the system fans and the lower operating temp of the direct water-cooled system components.

The direct energy savings at the enclosure level, combined with the potential for significant facility energy savings, makes the NVIDIA HGX Rubin NVL8 an excellent choice for customers that are burdened by high energy costs or with a sustainability mandate.

The liquid cooling solution consists of a coolant distribution unit (CDU), rear manifold, Disconnect Quick Connect (QD)s, and cold plates for the compute tray and all NVL Switch tray components.

The following table lists the reference design cooling percentages.

Water connections to the MGX Rack are provided using hoses that connect directly to the coolant distribution unit (CDU), inlet and return, either an in-rack CDU or in-row CDU. As shown in the following figure, hose lengths required depend on the placement of the Compute Trays and VR NVL Switch Trays in the rack cabinet.

The midplane in the chassis routes the water via QuickConnects to each of the Server Racks.

The VR offers two options for the rack level liquid cooling manifold: bottom feed and top feed. Both options are shown below.

Figure 10. Water connections to the MGX Rack cabinet

Water connections are made up of the following:

• Short hoses permanently attached to the rear of the rack cabinet
• Intermediate hoses, orderable as three different pairs of hoses, based on the location of the components in the rack.
• Connection to the CDU through additional hoses or valves on the data center water loop. Lenovo Datacenter Services can provide the end-to-end enablement for your water infrastructure needs.

For additional information, see the Cooling section.

To support the onsite setup for the direct water-cooled solution, a service kit is available providing a flow meter, bleed hose, pressure gauge and vent valve.

The liquid cooling manifold information is listed in the following table.

For details on Lenovo Neptune direct water-cooling standards, see the following document:

Lenovo Neptune Direct Water-Cooling Standards https://lenovopress.lenovo.com/lp2018-lenovo-neptune-direct-water-cooling-standards

The Lenovo MGX 48U Rack for HGX Rubin NVL8 Rack Solution supports up to 4 power shelves. The use of water-cooled power conversion stations enables an even greater amount of heat can be removed from the data center using water instead of air-conditioning.

The Power Shelves supply internal system power to a 54V busbar. This innovative design merges power conversion, rectification, and distribution into these power units, a departure from traditional setups that require separate units, including separate power supplies, resulting in best-in-class efficiency.

Redundancy

• 1 redundant power shelf with only 3 required for full rack power, allowing for 4-to-make-3 redundancy in the data center

The following table lists specifications for Power Shelf.

The following image shows the power shelves.

Figure 11. Power shelves

The following table lists Bus Bar specifications.

The following table lists other rack components.

The System Management Module includes the extender board with E1.S, the carrier board with BMC, HMC, TPM, utility ethernet and the front panel IO.

The NVIDIA management is the BMC board option which provides advanced control, monitoring, and alerting functions. The BMC is used in conjunction with the Host Module Controller (HMC) to provide OOB management. The HMC interposer board is within the GPU Compute Tray, which connects the HMC board and the BMC board to the VR HPM via cables.

The HMC and BMC work together to provide the out-of-band manageability for the tray.

Topics in this section:

• Local management
• NVIDIA AI Enterprise

Additionally, the following table lists the features of the BMC Server Management.

The following table lists the features of the SBIOS.

NVIDIA AI Enterprise

The NVIDIA HGX Rubin NVL8 is designed for NVIDIA AI Enterprise, which is a comprehensive suite of artificial intelligence and data analytics software designed for optimized development and deployment in enterprise settings.

NVIDIA AI Enterprise includes workload and infrastructure management software known as Base Command Manager. This software provisions the AI environment, incorporating the components such as the Operating System, Kubernetes (K8S), GPU Operator, and Network Operator to manage the AI workloads.

Additionally, NVIDIA AI Enterprise provides access to ready-to-use open-sourced containers and frameworks from NVIDIA like NVIDIA NeMo, NVIDIA RAPIDS, NVIDIA TAO Toolkit, NVIDIA TensorRT and NVIDIA Triton Inference Server.

• NVIDIA NeMo is an end-to-end framework for building, customizing, and deploying enterprise-grade generative AI models; NeMo lets organizations easily customize pretrained foundation models from NVIDIA and select community models for domain-specific use cases.
• NVIDIA RAPIDS is an open-source suite of GPU-accelerated data science and AI libraries with APIs that match the most popular open-source data tools. It accelerates performance by orders of magnitude at scale across data pipelines.
• NVIDIA TAO Toolkit simplifies model creation, training, and optimization with TensorFlow and PyTorch and it enables creating custom, production-ready AI models by fine-tuning NVIDIA pretrained models and large training datasets.
• NVIDIA TensorRT, an SDK for high-performance deep learning inference, includes a deep learning inference optimizer and runtime that delivers low latency and high throughput for inference applications. TensorRT is built on the NVIDIA CUDA parallel programming model and enables you to optimize inference using techniques such as quantization, layer and tensor fusion, kernel tuning, and others on NVIDIA GPUs. https://developer.nvidia.com/tensorrt-getting-started
• NVIDIA TensorRT-LLM is an open-source library that accelerates and optimizes inference performance of the latest large language models (LLMs). TensorRT-LLM wraps TensorRT’s deep learning compiler and includes optimized kernels from FasterTransformer, pre- and post-processing, and multi-GPU and multi-node communication. https://developer.nvidia.com/tensorrt
• NVIDIA Triton Inference Server optimizes the deployment of AI models at scale and in production for both neural networks and tree-based models on GPUs.

It also provides full access to the NVIDIA NGC catalogue, a collection of tested enterprise software, services and tools supporting end-to-end AI and digital twin workflows and can be integrated with MLOps platforms such as ClearML, Domino Data Lab, Run:ai, UbiOps, and Weights & Biases.

Finally, NVIDIA AI Enterprise introduced NVIDIA Inference Microservices (NIM), a set of performance-optimized, portable microservices designed to accelerate and simplify the deployment of AI models. Those containerized GPU-accelerated pretrained, fine-tuned, and customized models are ideally suited to be self-hosted and deployed on the NVIDIA HGX Rubin NVL8.

Topics in this section:

• Security features

Security features

The server offers the following electronic security features:

• Support for Platform Firmware Resiliency (PFR) hardware Root of Trust (RoT)
• Firmware signature processes compliant with FIPS and NIST requirements
• Administrator and power-on password
• Integrated Trusted Platform Module (TPM) supporting TPM 2.0

The server is NIST SP 800-147B compliant.

The following table lists the security options for the Lenovo NVIDIA HGX Rubin NVL8.

The Lenovo MGX 48U Rack for HGX Rubin NVL8 has the following dimensions:

• Width: 600 mm (23.6 inches)
• Height: 2,294 mm (90.3 inches)
• Depth: 1,068 mm (42.05 inches) - Expands to 1,200 mm with extension frame
• Weight: 185 kg (407.8 lbs)

Lenovo NVIDIA HGX Rubin NVL8 is installed onto the rack. Each Rack Server tray has the following dimensions:

• Width: 448 mm (17.64 inches)
• Height: 87.0 mm (3.43 inches)
• Depth: 800 mm (31.5 inches)
• Weight: 66 kg (145.5 lbs)

The Lenovo NVIDIA SN2201 48 port DC switch has the following dimensions:

• Width: 438 mm (17.3 inches)
• Height: 43.6 mm (1.72 inches)
• Depth: 781 mm (28.3 inches)
• Weight: 11.39 kg (25.1 lbs)

The Power Shelf has the following dimensions:

• Width: 448 mm (17.6 inches)
• Height: 131.4 mm (5.17 inches)
• Depth: 894.8 mm (35.23 inches)

The following table lists the detailed dimensions. See the figure below for the definition of each dimension.

Figure 13. Enclosure dimensions

TheLenovo NVIDIA HGX Rubin NVL HG658X V4 Rack Server is supported in the following environment.

Topics in this section:

• Air temperature and humidity
• Water requirements
• Particulate contamination

Air temperature and humidity

Air temperature/humidity requirements:

• Air Temperature
  • Operating supports ASHRAE A3
  • Non-operating: -40°C to 70°C
• Air Humidity
  • Operating: 8-85% RH
  • Non-op: 5-90% RH
• Air Altitude
  • Operating: <10 kft (3050 m)
  • Non-Op: <40 kft (12 km)

Water requirements

For details on the Lenovo NVIDIA HGX Rubin NVL8 Rack Server direct water-cooling standards, see the following values:

• Liquid Supply Temperature: 17°C to 45°C
• Liquid Return Temperature: <65°C
• Liquid burst Pressure: >217 psig
• Liquid operating Pressure: <72 psig
• Liquid filtration requirement: 25 um
• Use either PG25 or DiH20 w/ chem
• Wetted matls: follow ASHRAE approved list

The following specifications indicate the limits of particulates that the system can tolerate:

• Reactive gases:
  • The copper reactivity level shall be less than 200 Angstroms per month (Å/month)
  • The silver reactivity level shall be less than 200 Å/month
• Airborne particulates:
  • The room air should be continuously filtered with MERV 8 filters.
  • Air entering a data center should be filtered with MERV 11 or preferably MERV 13 filters.
  • The deliquescent relative humidity of the particulate contamination should be more than 60% RH
  • Environment must be free of zinc whiskers

For additional information, see the Specifications section of the documentation for the server, available from the Lenovo Documents site, https://pubs.lenovo.com/

The NVIDIA HGX Rubin NVL8 conforms to the following standards:

The Lenovo MGX 48U Rack for HGX Rubin NVL8 has the following warranty:

Lenovo MGX 48U Rack for HGX Rubin NVL8 Rack Solution (machine type 7DJV) rack and included components require a 3-Year Premier Next Business Day (NBD) Warranty.

• Note: Spare parts are required in order for NBD service.

• Support is provided at the rack level and the serial number of the rack is required for support entitlement.

Lenovo Infrastructure Services empower you at every stage of your IT lifecycle. From expert advisory and strategic planning to seamless deployment and ongoing support, we ensure your infrastructure is built for success. Our comprehensive services accelerate time to value, minimize downtime, and free your IT staff to focus on driving innovation and business growth.

In this section:

• Lenovo Advisory Services
• Lenovo Implementation Services
• Infrastructure Deployment Services
• Lenovo Support Services
• Lenovo Managed Services
• Lenovo Sustainability Services

Lenovo Advisory Services

Lenovo Advisory Services simplify the planning process, enabling customers to build future-proofed strategies in as little as six weeks. Consultants provide guidance on projects including VM migration, storage, backup and recovery, and cost management to accelerate time to value, improve cost efficiency, and build a flexibly scalable foundation.

• Assessment Services

  An Assessment helps solve your IT challenges through an onsite, multi-day session with a Lenovo technology expert. We perform a tools-based assessment which provides a comprehensive and thorough review of a company's environment and technology systems. In addition to the technology-based functional requirements, the consultant also discusses and records the non-functional business requirements, challenges, and constraints. Assessments help organizations like yours, no matter how large or small, get a better return on your IT investment and overcome challenges in the ever-changing technology landscape.

Lenovo Implementation Services

Optimize your IT operations by shifting labor-intensive functions to Lenovo's skilled technicians for seamless on-site or remote deployment, configuration, and migration. Enjoy peace of mind, faster time to value, and comprehensive knowledge sharing with your IT staff, backed by our best-practice methodology.

Infrastructure Configuration Services

• Factory Integrated Services

  A suite of value-added offerings provided during the manufacturing phase of a server or storage system that reduces time to value. These services aim at improving your hardware deployment experience and enhance the quality of a standard configuration before it arrives at your facility.

Infrastructure Deployment Services

Lenovo Infrastructure Deployment Services help you move from hardware delivery toward production with the planning, installation, configuration, integration, validation, documentation, and operational handoff support your project requires. Choose from clearly defined Premier Deploy for Infrastructure options, Specialized Deployment Services, or Hardware Installation Services based on your platform, project complexity, internal IT capabilities, and desired level of assurance.

For more information, consult our Lenovo Infrastructure Deployment Services Catalog.

• Premier Deploy for Infrastructure

  Clearly defined deployment options for eligible ThinkSystem servers, ThinkSystem storage arrays, and ThinkAgile solutions.

  Table 23. Premier Deploy for Infrastructure

  	Standard Deploy	Premier Deploy	Premier Deploy Plus
  Choose when:	You need straightforward physical hardware installation and customer acceptance.	You need operating system installation and/or system configuration, integration and validation.	You need extended post-deployment assurance and visibility after implementation.
  You gain:	An installed, accepted, and documented hardware foundation.	A validated, integrated environment ready to move into production.	Added assurance as your team transitions into ongoing operations.

  For more information, see the Premier Deploy for Infrastructure 1-pager.

• Specialized Deployment Services

  Focused deployment expertise for specific technologies or project requirements when Premier Deploy for Infrastructure is not supported or is not the right fit for your project. Services can support ThinkSystem storage arrays, ThinkAgile solutions, HPC, liquid cooling, and private cloud environments through activities such as readiness planning, configuration, integration, validation, documentation, knowledge transfer, and operational handoff, as applicable. Availability and scope vary by platform, project requirements, geography, and service scope.

• Hardware Installation Services

  Expert physical installation for eligible hardware outside the Premier Deploy for Infrastructure scope, or for projects that require installation only.

• Hardware Installation Plus

  Extends Hardware Installation Services with:

  • RAID configuration
  • Supported operating system installation
  • Driver-level verification

Lenovo Support Services

In addition to response time options for hardware parts, repairs, and labor, Lenovo offers a wide array of additional support services to ensure your business is positioned for success and longevity. Our goal is to reduce your capital outlays, mitigate your IT risks, and accelerate your time to productivity.

• Premier Support for Infrastructure

  Helps you build business resilience with expert-led support, responsive issue resolution, and coordinated troubleshooting across supported Lenovo server and storage environments. Direct access to Lenovo expertise and accountable end-to-end case ownership help reduce disruption and simplify ongoing infrastructure operations.

• Premier Support Plus for Infrastructure

  Gain all the benefits of Premier Support for Infrastructure, with proactive and predictive support capabilities, designated account oversight, preventative maintenance guidance, and Keep Your Drive to help identify potential issues earlier, plan service more effectively, and maintain control of sensitive data.

• Committed Service Repair (CSR)

  Our commitment to ensuring the fastest, most seamless resolution times for mission-critical systems that require immediate attention to ensure minimal downtime and risk for your business. This service is only available for machines under the Premier 4-Hour Response SLA.

• Multivendor Support Services (MVS)

  Your single point of accountability for resolution support across a vast range of leading Server, Storage, and Networking OEMs, allowing you to manage all your supported infrastructure devices seamlessly from a single source.

• Keep Your Drive (KYD)

  Protect sensitive data and maintain compliance with corporate retention and disposal policies to ensure your data is always under your control, regardless of the number of drives that are installed in your Lenovo server.

• Technical Account Manager (TAM)

  Your single point of contact to expedite service requests, provide status updates, and furnish reports to track incidents over time, ensuring smooth operations and optimized performance as your business grows.

• Enterprise Software Support (ESS)

  Gain comprehensive, single-source, and global support for a wide range of server operating systems and Microsoft server applications.

For more information, consult our Lenovo Operational Support Services for Data Centers Services Brochure.

Lenovo Managed Services

Achieve peak efficiency, high security, and minimal disruption with Lenovo's always-on Managed Services. Our real-time monitoring, 24x7 incident response, and problem resolution ensure your infrastructure operates seamlessly. With quarterly health checks for ongoing optimization and innovation, Lenovo's remote active monitoring boosts end-user experience and productivity by keeping your data center's hardware performing at its best.

• Managed Services

  Lenovo Managed Services provides continuous 24x7 remote monitoring (plus 24x7 call center availability) and proactive management of your data center using state-of-the-art tools, systems, and practices by a team of highly skilled and experienced Lenovo services professionals.

  Quarterly reviews check error logs, verify firmware & OS device driver levels, and software as needed. We’ll also maintain records of the latest patches, critical updates, and firmware levels, to ensure your systems are providing business value through optimized performance.

Lenovo Sustainability Services

Lenovo Sustainability Services helps you address sustainability priorities across the IT lifecycle, from offsetting emissions associated with device purchases and extending equipment life to securely managing assets at end of life. Together, these services support more circular IT practices while helping you protect data, manage costs, and recover value from technology investments.

• Asset Recovery Services

  Lenovo Asset Recovery Services (ARS) provides a secure, seamless solution for managing end-of-life IT assets, ensuring data is safely sanitized while contributing to a more circular IT lifecycle. By maximizing the reuse or responsible recycling of devices, ARS helps businesses meet sustainability goals while recovering potential value from their retired equipment. For more information, see the Simplify IT Asset Recovery page.

• CO2 Offset Services

  Lenovo’s CO2 Offset Services offer a simple and transparent way for businesses to take tangible action on their IT footprint. By integrating CO2 offsets directly into device purchases, customers can easily support verified climate projects and track their contributions, making meaningful progress toward their sustainability goals without added complexity.

• Lenovo Certified Refurbished

  Lenovo Certified Refurbished offers a cost-effective way to support IT circularity without compromising on quality and performance. Each device undergoes rigorous testing and certification, ensuring reliable performance and extending its lifecycle. With Lenovo’s trusted certification, you gain peace of mind while making a more sustainable IT choice.

• Data Center Power and Cooling Services

  The Data Center Infrastructure team will provide solution design and implementation services to support the power and cooling needs of the multi-node chassis and multi-rack solutions. This includes designing for various levels of power redundancy and integration into the customer power infrastructure. The Infrastructure team will work with site engineers to design an effective cooling strategy based on facility constraints or customer goals and optimize a cooling solution to ensure high efficiency and availability. The Infrastructure team will provide the detailed solution design and complete integration of the cooling solution into the customer data center. In addition, the Infrastructure team will provide rack and chassis level commissioning and stand-up of the water-cooled solution which includes setting and tuning of the flow rates based on water temperature and heat recovery targets. Lastly, the Infrastructure team will provide cooling solution optimization and performance validation to ensure the highest overall operational efficiency of the solution.

The Lenovo NVIDIA HGX Rubin NVL8 Rack Server, 8x NVIDIA Rubin GPUs, SN2201 Management Switch and Power Shelves are supported in the following racks:

• 7DJVCTO3WW Lenovo MGX 48U Rack for HGX Rubin NVL8

Considering the weight of the trays in the server, an onsite material lift is required to allow service by a single person. If you do not already have a material lift available, Lenovo offers the Genie Lift GL-8 material lift as configurable option to the rack cabinets. Ordering information is listed in the following table.

Note: If neither the Genie Lift GL-8 nor the ServerLift SL-350x is available onsite when onsite service is required, the customer will be responsible for getting the system to a suitable work surface (with access to power) prior to service technician arrival and returning the system to the rack when service is complete prior to service technician departure.

Why wait to obtain the technology you need now? No payments for 90 days and predictable, low monthly payments make it easy to budget for your Lenovo solution.

• Flexible

  Our in-depth knowledge of the products, services and various market segments allows us to offer greater flexibility in structures, documentation and end of lease options.

• 100% Solution Financing

  Financing your entire solution including hardware, software, and services, ensures more predictability in your project planning with fixed, manageable payments and low monthly payments.

• Device as a Service (DaaS)

  Leverage latest technology to advance your business. Customized solutions aligned to your needs. Flexibility to add equipment to support growth. Protect your technology with Lenovo's Premier Support service.

• 24/7 Asset management

  Manage your financed solutions with electronic access to your lease documents, payment histories, invoices and asset information.

• Fair Market Value (FMV) and $1 Purchase Option Leases

  Maximize your purchasing power with our lowest cost option. An FMV lease offers lower monthly payments than loans or lease-to-own financing. Think of an FMV lease as a rental. You have the flexibility at the end of the lease term to return the equipment, continue leasing it, or purchase it for the fair market value. In a $1 Out Purchase Option lease, you own the equipment. It is a good option when you are confident you will use the equipment for an extended period beyond the finance term. Both lease types have merits depending on your needs. We can help you determine which option will best meet your technological and budgetary goals.

Ask your Lenovo Financial Services representative about this promotion and how to submit a credit application. For the majority of credit applicants, we have enough information to deliver an instant decision and send a notification within minutes.

The following sales training courses are offered for employees and partners (login required). Courses are listed in date order.

1. Partner Technical Webinar - NVIDIA Software
   2026-10-27 | 60 minutes | Employees and Partners
   Details
   Partner Technical Webinar - NVIDIA Software
   
   

   In this 60-minute replay, there were two topics. First, Farah Fallah Toosi, Lenovo NVIDIA Software PM, reviewed the Lenovo NVIDIA Software portfolio positioning and how to attach. Next, Anh Duong, NVIDIA Solution Architect, reviewed Building Agents with NVIDIA Tools showing which NVIDIA Software products are used for Agentic AI.

   Published: 2026-10-27
   Length: 60 minutes
   

   Start the training:
   Employee link: Grow@Lenovo
   Partner link: Lenovo 360 Learning Center
   

   Course code: PTWOCT2326
2. From Use Case to AI Outcomes - Client Managers
   2026-10-14 | 20 minutes | Employees Only
   Details
   From Use Case to AI Outcomes - Client Managers
   
   

   As a Client Manager, you don’t need to design an AI Factory or answer every technical question. You do need to recognize when an initial request may point to something bigger, ask questions that uncover the business need, explain Lenovo and NVIDIA’s value at a high level and know when to brin in additional expertise.
   
   
   
   This course is about what happens when your existing AI knowledge meets a customer conversation. By completing it you prepare yourself to:
   
   - Explain how an AI Factory helps customers move from isolated AI efforts to repeatable, scalable business outcomes.
   
   - Describe, at a high level, how Lenovo, NVIDIA and Lenovo Services work together to support AI initiatives.
   
   - Recognize customer signals that may indicate a broader AI opportunity, beyond an initial request.
   
   - Identify business-focused questions that help clarify the customer’s goals, constraints and readiness.
   
   
   
   Revision Date: 10/14/2026

   Published: 2026-10-14
   Length: 20 minutes
   

   Start the training:
   Employee link: Grow@Lenovo
   

   Course code: DNVIB101
3. HPC VTT - Building a Converged HPC and AI Platform with Slurm and Slinky
   2026-10-13 | 55 minutes | Employees Only
   Details
   HPC VTT - Building a Converged HPC and AI Platform with Slurm and Slinky
   
   

   Watch this HPC VTT session to see Aurelien Ortiz, Software Architect HPC & AI at Lenovo, and Farah Fallah Toosi, NVIDIA AI Software Product Manager at Lenovo, as they discuss building a converged HPC and AI platform with Slurm and Slinky, and how to sell Slurm and Slinky.

   Published: 2026-10-13
   Length: 55 minutes
   

   Start the training:
   Employee link: Grow@Lenovo
   

   Course code: DVHP239
4. Partner Technical Webinar - AI Stories - Part 1
   2026-09-04 | 60 minutes | Employees and Partners
   Details
   Partner Technical Webinar - AI Stories - Part 1
   
   

   In this 60 minute replay, John Encizo’s AI team joined us to share some AI Stories: from Requirements to Architecture. Part 1

   Published: 2026-09-04
   Length: 60 minutes
   

   Start the training:
   Employee link: Grow@Lenovo
   Partner link: Lenovo 360 Learning Center
   

   Course code: AUG2826
5. Think AI Weekly: NVIDIA AI Software Use Cases
   2026-08-13 | 23 minutes | Employees Only
   Details
   Think AI Weekly: NVIDIA AI Software Use Cases
   
   

   Please watch this session of Think AI Weekly as Farah Fallah Toosi, NVIDIA AI Software Product Manager talks to us about NVIDIA AI Software.
   
   Topics include:
   
   -Understand key NVIDIA AI software
   
   - NemoClaw
   
   - Confidential Computing
   
   - NVIDIA AI Enterprise (changes)
   
   -Connect use cases to customer needs
   
   -Position NVIDIA software value

   Published: 2026-08-13
   Length: 23 minutes
   

   Start the training:
   Employee link: Grow@Lenovo
   

   Course code: DTAIW125
6. Think AI Weekly: NVIDIA Software Overview
   2026-07-14 | 24 minutes | Employees Only
   Details
   Think AI Weekly: NVIDIA Software Overview
   
   

   Watch this Think AI Weekly video to see Farah Fallah Toosi introduce the suite of NVIDIA software available to sellers. Our second speaker, David Watts, covered Lenovo Press and the information it offers, and highlighted where to find training materials linked to product guides.

   Published: 2026-07-14
   Length: 24 minutes
   

   Start the training:
   Employee link: Grow@Lenovo
   

   Course code: DTAIW179
7. HPC VTT: Unlocking Hybrid HPC + AI_ Slinky Bridge for Unified GPU Workloads
   2026-03-10 | 62 minutes | Employees Only
   Details
   HPC VTT: Unlocking Hybrid HPC + AI_ Slinky Bridge for Unified GPU Workloads
   
   

   View this session to hear from our speakers, Aurelien Ortiz, Software Architect
   HPC & AI, Lenovo and Nick Ihli, Sr Product Manager System Software – Slurm, SchedMD, as they explain how Slinky helps with unified GPU workloads. Topics include:
   Why Slinky? Bridging HPC and Cloud-Native AI Workloads
   How Slinky Bridge Works: Architecture, Components, and Flow
   Deployment Requirements & How to Position Slinky

   Published: 2026-03-10
   Length: 62 minutes
   

   Start the training:
   Employee link: Grow@Lenovo
   

   Course code: DVHPC231
8. Edge VTT - NVIDIA Robotics Platform
   2026-01-08 | 67 minutes | Employees Only
   Details
   Edge VTT - NVIDIA Robotics Platform
   
   

   In this session we feature speakers from both NVIDIA and Lenovo. Attendees will learn about NVIDIA’s platform stack for Robotics and what Lenovo is doing in the field of robotics.
   
   During this session we will dive into NVIDIA’s three-computer stack for Physical AI. Our speaker will explore libraries and workflows to develop, train, simulate, deploy, operate, and optimize AI robot systems and software. This session will cover the basics of the technical platform, how to get started and case studies from some NVIDIA's ecosystem.
   
   Objectives:
   Discuss acceleration libraries
   Describe simulation workflows
   List foundational models for robotics
   
   Tags: Artificial Intelligence (AI), Sales, Software Platforms, Technical Sales

   Published: 2026-01-08
   Length: 67 minutes
   

   Start the training:
   Employee link: Grow@Lenovo
   

   Course code: DVEDG223
9. Lenovo Unlocks the Power of NIM: Overview, Industry Use Cases and Lenovo Services
   2025-10-16 | 55 minutes | Employees and Partners
   Details
   Lenovo Unlocks the Power of NIM: Overview, Industry Use Cases and Lenovo Services
   
   

   Join us for an insightful session with our Lenovo speakers, Farah Toos and Dinesh Tripathi where we’ll explore the transformative potential of NVIDIA Inference Microservices (NIM). This webinar will provide a comprehensive overview of NIMs, highlighting how it streamlines operations, enhances scalability, and drives innovation across industries.
   
   Discover real-world use cases in sectors such as healthcare, manufacturing, retail, and finance, and learn how Lenovo’s portfolio of services—including deployment, optimization, and lifecycle support—can help your customers maximize the value of their infrastructure investments.
   Whether you're engaging with enterprise clients or mid-market opportunities, this session will equip you with the knowledge and tools to position Lenovo’s NIM solutions effectively and drive impactful conversations.
   
   Key Takeaways:
   - Understand the core capabilities and benefits of NIM
   - Explore industry-specific applications and success stories
   - Learn how Lenovo services complement and enhance NIM deployments
   - Gain selling strategies and resources to support customer engagements
   
   Tags: Artificial Intelligence (AI), NVIDIA,Services,Technical Sales, ThinkAgile, ThinkSystem

   Published: 2025-10-16
   Length: 55 minutes
   

   Start the training:
   Employee link: Grow@Lenovo
   Partner link: Lenovo 360 Learning Center
   

   Course code: DVCLD228
10. AI VTT: NVIDIA Run:ai
    2025-07-02 | 75 minutes | Employees Only
    Details
    AI VTT: NVIDIA Run:ai
    
    

    NVIDIA Run:ai is a GPU orchestration and optimization platform designed to help organizations maximize their GPU compute resources for AI workloads. It accelerates AI development, reduces costs, and improves AI development cycles by enabling dynamic allocation and scheduling of GPU resources, as well as workload submission and sharing. Essentially, it provides a centralized interface to manage AI compute infrastructure, making it easier for AI teams to access and utilize GPUs effectively.
    
    Join Carlos Huescas from Lenovo, Sandeep Brahmarouthu and Robert Magno from NVIDIA as they discuss NVIDIA Run:ai. Topics include:
    •What is Run:ai and its capabilities?
    •Customer segmentation for Run:ai
    •How to order, part numbers and licensing
    •Demo of Run:ai
    
    Tags: Artificial Intelligence (AI), NVIDIA

    Published: 2025-07-02
    Length: 75 minutes
    

    Start the training:
    Employee link: Grow@Lenovo
    

    Course code: DVAI218
11. VTT Edge: Understanding Visual AI Agents with NVIDIA June 2025
    2025-06-16 | 60 minutes | Employees and Partners
    Details
    VTT Edge: Understanding Visual AI Agents with NVIDIA June 2025
    
    

    Join our guest speakers from NVIDIA as they discuss what’s behind the scenes of visual AI Agents for Smart Cities, Smart Spaces and Manufacturing. Explore the modular approach to building a workforce of AI Agents. Topics include:
    
    • Sensors which feed the AI Agents
    • How AI agents improve safety and prevent accidents in Smart Spaces
    • Demo: Modular development of AI Agents
    
    Tags: Artificial Intelligence (AI), Technical Sales, NVIDIA

    Published: 2025-06-16
    Length: 60 minutes
    

    Start the training:
    Employee link: Grow@Lenovo
    Partner link: Lenovo 360 Learning Center
    

    Course code: DVEDG221
12. Lenovo Cloud Architecture VTT: Supercharge Your Enterprise AI with NVIDIA AI Enterprise on Lenovo Hybrid AI Platform
    2025-04-17 | 75 minutes | Employees and Partners
    Details
    Lenovo Cloud Architecture VTT: Supercharge Your Enterprise AI with NVIDIA AI Enterprise on Lenovo Hybrid AI Platform
    
    

    Join us for an in-depth webinar with Justin King, Principal Product Marketing Manager for Enterprise AI exploring the power of NVIDIA AI Enterprise, delivering Generative and Agentic AI outcomes deployed with Lenovo Hybrid AI platform environments.
    In today’s data-driven landscape, AI is evolving at high speed, with new techniques delivering more accurate responses. Enterprises are seeking not just an understanding but also how they can achieve AI-driven business outcomes.
    With this, the demand for secure, scalable, and high-performing AI operations-and the skills to deliver them-is top of mind for many. Learn how NVIDIA AI Enterprise, a comprehensive software suite optimized for NVIDIA GPUs, provides the tools and frameworks, including NVIDIA NIM, NeMo, and Blueprints, to accelerate AI development and deployment while reducing risk-all within the control and security of your Lenovo customer’s hybrid AI environment.
    
    Tags: Artificial Intelligence (AI), Cloud, Data Management, Nvidia, Technical Sales

    Published: 2025-04-17
    Length: 75 minutes
    

    Start the training:
    Employee link: Grow@Lenovo
    Partner link: Lenovo 360 Learning Center
    

    Course code: DVCLD221
13. AI VTT: GTC Update and The Lenovo LLM Sizing Guide
    2025-03-12 | 86 minutes | Employees Only
    Details
    AI VTT: GTC Update and The Lenovo LLM Sizing Guide
    
    

    Please view this session that is two parts. Part one is Robert Daigle, Director, Global AI Solutions and Hande Sahin-Bahceci, AI Solutions Marketing Leader explaining the upcoming announcements for NVIDIA GTC. Part Two is Sachin Wani, AI Data Scientist explaining the Lenovo LLM Sizing Guide with these topics:
    
    • Minimum GPU requirements for fine-tuning/training and inference
    • Gathering requirements for the customer's use case
    • LLMs from a technical perspective
    
    Tags: Artificial Intelligence (AI), Technical Sales

    Published: 2025-03-12
    Length: 86 minutes
    

    Start the training:
    Employee link: Grow@Lenovo
    

    Course code: DVAI214
14. VTT AI: Components of the AI Stack and Where Lenovo Sits November 2024
    2024-11-26 | 75 minutes | Employees Only
    Details
    VTT AI: Components of the AI Stack and Where Lenovo Sits November 2024
    
    

    Join Per Ljungstrom, Lenovo Principal TC EMEA, as he explores AI concepts where innovations meet simplified predefined solutions which deploy at scale. Topics for this session include:
    • Associating software with the ground level of hardware
    • Attach NVIDIA AI Enterprise, Microsoft, Tiber AI Stacks and more
    • AI at the Edge and the complete solution
    • What to consider when talking AI Stack with your customer
    
    Tags: Artificial Intelligence (AI), Cloud, Technical Sales, Technology solutions, ThinkEdge

    Published: 2024-11-26
    Length: 75 minutes
    

    Start the training:
    Employee link: Grow@Lenovo
    

    Course code: DVAI210
15. VTT AI: NVIDIA OVX
    2024-10-23 | 55 minutes | Employees and Partners
    Details
    VTT AI: NVIDIA OVX
    
    

    Please join this session as Steven Puzio, Global Sales Leader of NVIDIA Omniverse speaks to us about these topics:
    
    • OVX use cases
    • Target customers
    • OVX reference architectures
    • Parts, pieces and technical details
    
    Tags: Artificial Intelligence (AI), Nvidia

    Published: 2024-10-23
    Length: 55 minutes
    

    Start the training:
    Employee link: Grow@Lenovo
    Partner link: Lenovo 360 Learning Center
    

    Course code: DVAI209
16. Think AI Weekly: Ride the NVIDIA Wave for AI
    2024-10-07 | 60 minutes | Employees Only
    Details
    Think AI Weekly: Ride the NVIDIA Wave for AI
    
    

    In this session, a panel including speakers from NVIDIA, Lenovo IDG and Lenovo ISG address the topics:
    •Leveraging AI workstations to start an AI journey
    •Leading an ISG sale with NVIDIA AI Enterprise
    •NVIDIA sales tools available for Lenovo sellers
    •NVIDIA training on grow@lenovo and more
    
    Tags: Artificial Intelligence (AI), Nvidia

    Published: 2024-10-07
    Length: 60 minutes
    

    Start the training:
    Employee link: Grow@Lenovo
    

    Course code: DTAIW121
17. Lenovo VTT Cloud Architecture - Unlock Gen AI with VMware Private AI Foundation with NVIDIA
    2024-07-16 | 60 minutes | Employees and Partners
    Details
    Lenovo VTT Cloud Architecture - Unlock Gen AI with VMware Private AI Foundation with NVIDIA
    
    

    In today's rapidly evolving digital landscape, businesses are hungry for the transformative power of Artificial Intelligence (AI). They see AI as the key to streamlining operations and unlocking exciting new opportunities. However, widespread adoption has been hampered by concerns surrounding privacy, the complexity of implementation, and the hefty costs associated with deploying and managing AI solutions at an enterprise level.
    Join Chris Gully and Baker Hull, Solutions Architects from VMware by Broadcom, as they discuss how Lenovo, NVIDIA, and VMware By Broadcom are partnering to deliver a private, secure, scalable, and flexible AI infrastructure solution that helps enterprise customers build and deploy AI workloads within their own private cloud infrastructure, ensure the control of sensitive data and compliance with regulatory requirements, ultimately driving faster time to value and achieving their AI objectives.
    
    Tags: Artificial Intelligence (AI), Cloud, Nvidia, ThinkAgile, VMware

    Published: 2024-07-16
    Length: 60 minutes
    

    Start the training:
    Employee link: Grow@Lenovo
    Partner link: Lenovo 360 Learning Center
    

    Course code: DVCLD214
18. Guidance for Selling NVIDIA Products at Lenovo for ISG
    2024-07-01 | 25 minutes | Employees and Partners
    Details
    Guidance for Selling NVIDIA Products at Lenovo for ISG
    
    

    This course gives key talking points about the Lenovo and NVIDIA partnership in the Data Center. Details are included on where to find the products that are included in the partnership and what to do if NVIDIA products are needed that are not included in the partnership. Contact information is included if help is needed in choosing which product is best for your customer. At the end of this session sellers should be able to explain the Lenovo and NVIDIA partnership, describe the products Lenovo can sell through the partnership with NVIDIA, help a customer purchase other NVIDIA product, and get assistance with choosing NVIDIA products to fit customer needs.
    
    Tags: Artificial Intelligence (AI), Nvidia

    Published: 2024-07-01
    Length: 25 minutes
    

    Start the training:
    Employee link: Grow@Lenovo
    Partner link: Lenovo 360 Learning Center
    

    Course code: DNVIS102

1-10 of 18 courses. Show All

For more information, see these resources:

• Lenovo NVIDIA HGX Rubin NVL8 server product page
  https://www.lenovo.com/us/en/p/servers-storage/servers/rack-scale-ai/lenovo-nvidia-vera-rubin-nvl72/len21ts0055
• Lenovo NVIDIA HGX Rubin NVL8 datasheet
• https://lenovopress.lenovo.com/ds0216
• Interactive 3D Tour of the Lenovo NVIDIA HGX Rubin NVL8:
• https://lenovopress.lenovo.com/lp2560
• Lenovo NVIDIA HGX Rubin NVL8 drivers and support
  https://datacentersupport.lenovo.com/us/en/products/solutions-and-software/aiinfrastructure/gb300nvl72/7djv/downloads/driver-list/

• ServerProven hardware compatibility:
  https://serverproven.lenovo.com/
• Data Center Solution Configurator (DCSC)
  https://dcsc.lenovo.com
• Lenovo Cluster solutions configurator (x-config)
  https://lesc.lenovo.com/products/hardware/configurator/worldwide/bhui/asit/index.html