Lenovo Servers

Enterprise-Grade Power for AI & HPC.

The ThinkSystem SR675 V3 is built to accelerate AI, HPC, and graphics-intensive workloads, delivering enterprise-grade performance with ultimate flexibility.

• Supports up to 8 GPUs including NVIDIA HGX™ H200 & AMD Instinct™
• Modular design with 4th/5th Gen AMD EPYC™ processors
• Neptune™ hybrid liquid cooling for efficiency & quiet performance
• Scales seamlessly with the Lenovo HPC & AI Software Stack
• Simplified management with LiCO web portal & open-source integration

ThinkSystem SR675 V3 — Versatility, scalability, and performance for the AI era.

Data sheet(s)

Lenovo HPC, Servers

Massive GPU Power for AI & HPC.

The ThinkSystem SR680a V3 delivers unmatched acceleration for AI, HPC, and simulation workloads, powered by dual 5th Gen Intel® Xeon® processors and up to 8 NVIDIA H100/H200/B200 GPUs.

• Enterprise-grade platform for AI & HPC at scale
• 8 GPUs with NVLink + high-speed interconnects
• 4TB DDR5 memory & up to 16 NVMe drives
• Flexible GPU choices to meet today’s and tomorrow’s workloads
• Backed by Lenovo AI portfolio

ThinkSystem SR680a V3 — Maximum acceleration for the AI era.

Data sheet(s)

Lenovo Servers

Accelerated Compute for AI & HPC.

The ThinkSystem SR685a V3, powered by dual 5th/4th Gen AMD EPYC™ processors and up to 8 fully connected GPUs, delivers extreme performance for the most demanding AI and HPC workloads.

• Choice of AMD Instinct™ MI300X or NVIDIA H100/H200 GPUs
• Up to 192GB HBM3 memory per GPU with ultra-fast interconnects
• 24 DDR5 DIMMs and flexible, open architecture design
• Simplified, resilient management with Lenovo XClarity
• Backed by Lenovo’s AI portfolio

ThinkSystem SR685a V3 — Maximum acceleration, ultimate flexibility.

Data sheet(s)

Lenovo HPC, Infrastructure, Network, Professional Services, Servers, Storage

OCF will leverage our deep expertise to design, deliver, deploy and support your next (or first) turnkey High Performance Computing (HPC) cluster based on Lenovo technology.

Every bespoke OCF / Lenovo cluster is built on proven technologies designed to maximise performance and optimise datacentre efficiency, tailored to your workloads and budget. Our solution design process is driven by a dedicated team of Presales Technical Architects who specialise in creating tailored HPC and storage architectures.

This team works in close collaboration with customers, technical partners and our in-house engineering teams to translate requirements into optimised, future-ready solutions. In addition to a broad market view of the industry and relevant technologies, the team is fully accredited and extremely experienced in architecting solutions based on Lenovo technologies.

For more information, read the PDF datasheet linked below or the CASE STUDY about Nottingham University.

Data sheet(s)

NVIDIA Servers

Powered by H200 multi-precision Tensor Cores, the eight-way HGX H200 unleashes over 32 peta-FLOPS of FP8 deep learning performance and more than 1TB of lightning-fast HBM memory which delivers unmatched speed and efficiency for the most demanding highest performance for generative AI and HPC workloads

The H200 SMX5 format with NVLink high-speed interconnects between the GPUs designed for dense GPU-to-GPU interconnects with NVIDIA HGX 200 partner and NVIDIA certified and approved servers with 4 or 8 GPUs.

Lenovo Servers

The Lenovo ThinkSystem SC777 V4 Neptune® Server represents the pinnacle of high-performance computing (HPC) and artificial intelligence (AI) infrastructure.

Based on the NVIDIA GB200 NVL4 platform, the ThinkSystem SC777 V4 server combines the latest high-performance NVIDIA GPU and Lenovo's market-leading water-cooling solution, which results in extreme HPC/AI performance in dense packaging. It integrates two NVIDIA Grace processors and four Blackwell B200 GPUs into a single, high-density server tray making outstanding use of an architecture that is purpose-built to handle the most demanding modern workloads, from large-scale AI model training and intensive scientific simulations to complex financial modelling and life sciences research.

This Lenovo SC777 V4 has been created using sixth generation Lenovo Neptune® direct water cooling platform. It uses 100% direct warm-water cooling, allowing the system to operate with inlet temperatures as high as 45°C (113°F), which effectively eliminates the need for energy-intensive mechanical chillers. The revolutionary design not only allows the processors and GPUs to run at peak performance without thermal throttling but also enables data centres to achieve up to a 40% reduction in energy costs. The captured heat can even be repurposed for building heating, providing a sustainable path toward exascale computing.

The SC777 V4 is designed for extreme density and seamless scalability in a Lenovo ThinkSystem N1380 Neptune enclosure. This 13U chassis houses up to eight SC777 V4 trays vertically, featuring patented blind-mate, drip-less connectors. This will ensure simple serviceability and maximum reliability. It can support up to 960GB of high-speed LPDR5 ECC memory and multiple PCIe Gen 5 expansion slots providing the massive I/O bandwidth necessary to support 400GbE and 800GbE networking to render data bottlenecks as a thing of the past in high-speed cluster environments.

Beyond its raw hardware capabilities, the Lenovo ThinkStation SC777 V4 offers a robust management and security framework through an XClarity Controller 3 (XCC3) and the Lenovo HPC & AI Software Stack. These tools provide administrators with a fully tested, open-source environment for effortless cluster orchestration and workload management. With integrated security features like TPM 2.0 and Secure Boot, OCF customers can deploy these systems with confidence, knowing they are investing in a secure, future-proof platform that bridges the gap between departmental computing and the world’s most powerful supercomputers.

Data sheet(s)

Xinnor Servers, Storage

xiNAS is a high-performance NFS solution powered by Xinnor's xiRAID engine, delivering throughput that reaches the practical limits of modern storage hardware. Explicitly designed for AI and HPC workloads, xiNAS keeps DGX/HGX systems continuously busy with optimized data delivery.

The solution delivers 5-10x higher performance than standard NFS implementations through a carefully optimized stack that includes tuned filesystem settings to avoid write amplification, optimized NFS server and client configurations, and smart network settings. This architecture enables more than 95% GPU utilization in AI training environments while remaining costeffective for midrange installations.

Introducing xiRAID — High-Performance Software RAID

Data sheet(s)

Galene AI, Servers

Galene rack-scale private AI server for organisations that need resilient multi-node operations and room to expand for new teams. The Dell PowerEdge R770 model with 2 x RTX Pro 6000 SE have 18 parallel interfaces perfect for around 180 users while the larger Dell PowerEdge XE7745 solution has 4 x RTX 6000 Pro SE which delivers 36 parallel interfaces and makes it perfect for around 350 users.

These Galene AI servers are enterprise-ready as:

• They are designed to maximise Return on Investment by enabling infinite adoption cases and projects
• They have unlimited scalability and are designed for multi-node enterprise scenarios
• They provide a fully proprietary and private solution that guarantees a clear Capax Investment.