OEM fiber optic solutions for data centers and telecom
Custom cabling and industrial communication modules

Where is the GB200 optical module used

Where is the GB200 optical module used

The NVIDIA GB200 uses 800G and 1.6T OSFP optical modules for high-speed inter-cabinet connections, complementing internal copper NVLink interconnects.Overview of Optical Module UsageIn the GB200 NVL72 rack-scale system, internal communication within a cabinet relies primarily on copper interconnects such as 800G/1600G DAC and ACC cables, which provide low-latency, high-bandwidth connections for up to 72 Blackwell GPUs and 36 Grace CPUs . For inter-cabinet or cluster-level connections, the system employs 800G and 1.6T OSFP optical modules, enabling scalable high-speed networking across multiple racks .800G Optical ModulesApplication: Used extensively in large GB200 clusters, such as a 576-GPU deployment, requiring tens of thousands of 800G DR8 optical modules .Function: Facilitate high-bandwidth interconnects for AI workloads, supporting NVLink and InfiniBand networks.Manufacturers: Eoptolink and InnoLight dominate the supply, with Eoptolink supplying over 60% of 800G modules for NVIDIA GB200 servers .1.6T Optical ModulesApplication: Deployed in dual-layer and three-layer InfiniBand networks for GB200 NVL72 systems, with a GPU-to-transceiver ratio of 1:2 or 1:3 depending on network topology .Function: Provide ultra-high-speed interconnects for inter-rack communication, supporting generative AI and large language model workloads.Technology: Utilize 200G PAM4 signaling across 8 parallel channels, achieving 1.6Tbps total transmission per module .Integration with GB200 ArchitectureInternal NVLink Network: Within a single GB200 cabinet, NVLink 5.0 provides up to 1.8TB/s bidirectional bandwidth per GPU, connected via copper DAC/ACC cables .External Cluster Connectivity: Optical modules extend the network beyond a single cabinet, enabling clusters of up to 576 GPUs with high-speed, low-latency interconnects .Rack-Scale Design: Each NVL72 rack includes compute trays, NVLink switch trays, and TOR switches, with optical modules interfacing at the top-of-rack or inter-rack level to maintain scalable AI data center performance .SummaryThe GB200 system balances copper and optical interconnects: copper DAC/ACC cables handle short-range, high-bandwidth internal connections, while 800G and 1.6T OSFP optical modules provide scalable, high-speed inter-cabinet connectivity. This combination ensures low latency, high throughput, and energy-efficient performance for AI workloads and large-scale GPU clusters .

NVIDIA GB200 Interconnect Architecture Analysis:

Explore the intricate interconnect architecture of the NVIDIA GB200, including NVLink bandwidth calculation, NVLINK 5.0 and 4.0 interconnect

AI Data Center Networking: How GPU Clusters Are Changing

⚠️ LPO (Linear Pluggable Optics): The AI Fabric Power Game-Changer LPO eliminates the DSP retimer chip inside the optical module, reducing power consumption by ~50% compared to

The Complete Optical Transceiver Procurement Landscape of

Each is on a completely different technology path, each uses a completely different procurement model, and each exposes Chinese optical module vendors in a completely different way.

NVIDIA GB200 Interconnect Architecture Analysis

In this process, high-speed interconnect devices such as optical modules will continue to play a crucial role in providing higher bandwidth and

GB200 Hardware Architecture

Today we are going to go from A to Z on the different form factors of GB200 and how they changed versus the prior 8 GPU HGX baseboard servers.

NVIDIA DGX GB Rack Scale Systems User Guide

This document contains detailed information about the hardware and software stack of DGX Grace Blackwell rack scale systems (such as GB200 and GB300). Each rack is an NVL72 rack (72-GPU

NVIDIA GB200 NVL72: PCB & System Architecture Explained

The GB200 Superchip module is not a standard PCB add-in card. It is a multi-chip module (MCM) that mounts to the compute tray baseboard via a high-density mezzanine connector.

What chips are used in optical modules? | Weyland

Optical modules are not isolated chips. Instead, they are complex systems composed of multiple high-speed electrical ICs, optoelectronic devices, and control chips.

NVIDIA DGX GB200 USER MANUAL Pdf Download | ManualsLib

Hardware The DGX GB200 is a rack scale solution for graphics processing units (GPUs) connected by an NVLink through the NVLink passive copper cable cartridge backplane.

NVIDIA GB200: Interconnect Architecture and Evolution

Each GB200 subsystem has 2 * 18 = 36 NVLink5 Ports. The external interconnect of the system does not use OSFP optical modules, but rather a

800G Data Center Interconnect Guide: DAC, AEC, AOC

DAC · ACC · AEC · AOC · Optical Transceivers — the complete engineer''s framework for choosing the right interconnect for every link in your AI

Semiconductor Fab Map — 65 Global Fabs with Capacity & Node Data

Explore 65 semiconductor fabs across 10 countries. Filter by foundry, process node, and production status. TSMC, Samsung, Intel capacity data.

Nvidia''s AI Supercomputer Costs $3 Million. To

The DGX GB200 NVL72 is an absolute monster in terms of power, price, and energy consumption. Several surprises are inside the cabinet—one of

How NVIDIA GB200 Utilizes 800G/1.6T DAC/ACC

Scalability for Large Clusters: While copper excels within cabinets, 800G/1600G optical modules support inter-cabinet connections, enabling

NVIDIA GB200 Supply Chain: The Global Ecosystem Explained

NVIDIA can sell GB200 modules to Foxconn for one customer, while directly supplying Dell for another. Some OEMs like Dell/HPE might design their own PCBs and simply purchase Blackwell GPUs from

directory-list-2.4.txt/directory-list-2.4.txt at main

Customer stories Events & webinars Ebooks & reports Business insights GitHub Skills

Nvidia''s AI Supercomputer Costs $3 Million. To

To Operate, It Uses a Switch With Nearly 2 Miles of Cable The DGX GB200 NVL72 is an absolute monster in terms of power, price, and energy

Deep Dive: Optical Module Market

Optical modules are typically defined by their bandwidth capacity, with the highest currently available being 800G modules. Next-generation 1.6T modules are expected to begin

Nvidia Vera Rubin: Data Center Impact Explained | ModulEdge

Nvidia Vera Rubin: 190-230 kW VR200 racks in H2 2026, 600 kW Kyber in 2027, 800 VDC power, all-liquid cooling, and stranded-facility risk.

NVIDIA GB200 NVL72 by HPE QuickSpecs | HPE

The NVIDIA GB200 NVL72 by HPE is a rack-scale system optimized to accelerate large AI model training, fine-tuning, and real-time inferencing for AI models over 1 trillion parameters.

NVIDIA DGX GB Rack Scale Systems User Guide

NVIDIA DGX GB Rack Scale Systems User Guide # The NVIDIA DGX GB Rack Scale Systems User Guide is also available as a PDF. This document contains detailed information about

NVIDIA B300 (Blackwell Ultra) Specs, Pricing, and Benchmarks (2026)

NVIDIA B300 specs vs B200, H200, H100. Live on-demand cloud pricing, infrastructure requirements, and which workloads justify the 288GB Blackwell Ultra.

Data Centre Cabling Standards 2026: TIA-942 vs BICSI

Data Center Cabling: Standards Guide, BICSI vs TIA-942, and AI-Era Strategy for 800G/1.6T (2026) A single AI GPU rack running NVIDIA''s

Introduction to NVIDIA GB200 Superchip and Liquid

Introduction The NVIDIA GB200 is a highly integrated supercomputing module based on NVIDIA''s Blackwell architecture. This module

Hardware — NVIDIA DGX GB Rack Scale Systems

Multiple power shelves are used in the rack to supply redundant power. The rack power consumption is approximately 120kW. The power shelf

Unveiling NVIDIA GB200 Interconnect Architecture

A single GB200 subsystem contains 2 * 18 = 36 NVLink5 Ports. The external interconnection of the entire system does not adopt OSFP optical

GB200 Hardware Architecture

GB200 Hardware Architecture - Component Supply Chain & BOM Hyperscale customization, NVLink Backplane, NVL36, NVL72, NVL576, PCIe

Nvidia DGX GB200 Manual | ManualsLib

As you insert the new module, press the latch button to ensure the lever stays in the open position. Fully insert the drive module and close the lever to lock it in place.

Why 1.6T Optical Transceivers Overtake 800G in 2026 AI Clusters?

1.6T optical transceivers are overtaking 800G in 2026 AI clusters because NVIDIA Blackwell GB200 systems require doubled bandwidth per port, Silicon Photonics technology reduces power

Co Packaged Optics (CPO) – Scaling with Light for the Next Wave of

Co-Packaged Optics (CPO) has long promised to transform datacenter connectivity, but it has taken a long time for the

More industry information

Contact Us

We Look Forward to Working with You

Contact Information

Phone +44 20 7946 0958
Address 1 Cornhill, London EC3V 3ND, United Kingdom

Send an Inquiry