Multiplexed Optical Transceivers for Low-Power Blade Server Fabrics

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Blade server chassis face bandwidth oversubscription and power consumption issues in Clos switched fabrics, and existing Ethernet-based data packet fabrics are not optimal for diverse computing applications.

Innovation Solution

Implementing multiplexed optical transceivers in a server chassis to establish a fabric topology that interconnects blade servers and a dedicated switch module, supporting both Ethernet and PCIe interfaces, while aggregating chassis management packet traffic.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If a Clos switched fabric is used to interconnect blade servers, then connectivity between blade servers and switch module is improved, but bandwidth oversubscription and power consumption increase

Engineering Contradiction:

ImproveconnectivityVSAvoidpower consumption

Core Design Contradiction:

ReliabilityVSUse of energy by moving object

Solution Approach 1:

The patent extracts the switching function from traditional Ethernet switches and relocates it to the fabric cards themselves. Each fabric card contains switching logic that enables direct peer-to-peer communication between blade servers, eliminating the need for centralized switch ASICs and reducing overall power consumption while maintaining full connectivity

Inventive Principle:

Principle #2Taking out (Extraction)

Solution Approach 2:

The switching fabric is segmented into distributed switching units embedded in each fabric card. Instead of a centralized switching architecture, each card independently handles routing decisions, dividing the switching workload and eliminating bandwidth oversubscription bottlenecks while reducing power consumption through localized intelligence

Inventive Principle:

Principle #1Segmentation

2Productivity

If more blade servers are added to increase computing capacity, then productivity improves, but bandwidth oversubscription and latency increase

Engineering Contradiction:

Improvecomputing capacityVSAvoidlatency

Core Design Contradiction:

ProductivityVSLoss of time

Solution Approach 1:

The network fabric is segmented into multiple independent routing paths through distributed switching units on each fabric card. This segmentation enables parallel data flows between blade servers, allowing computing capacity to scale without increasing latency as more servers are added to the chassis

Inventive Principle:

Principle #1Segmentation

3Ease of manufacture

If traditional switch ASIC hardware design is used, then Ethernet packet transport is standardized, but adaptability to diverse computing applications is limited

Engineering Contradiction:

ImprovestandardizationVSAvoidapplication compatibility

Core Design Contradiction:

Ease of manufactureVSAdaptability or versatility

Solution Approach 1:

The fabric cards are designed with universal interfaces and programmable switching logic that can handle multiple protocols and data types beyond traditional Ethernet. The architecture supports diverse computing applications including accelerated computing, distributed memory fabrics, and composable architectures through a single multi-functional platform

Inventive Principle:

Principle #6Universality (Multi-functionality)

Solution Approach 2:

The switching fabric implements dynamic, reconfigurable connections through programmable fabric cards rather than fixed ASIC routing. This allows the network topology and routing paths to be dynamically adjusted based on application requirements, providing adaptability to diverse computing workloads while maintaining ease of manufacture through standardized card form factors

Inventive Principle:

Principle #15Dynamics

Applied Scientific Principles

This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.

Function Achieved in This Case

Enhances bandwidth utilization and reduces power consumption by providing a flexible, efficient data fabric that supports diverse computing applications without the limitations of traditional switch ASIC hardware designs.

Implementation Method 1

multiplexed optical transceivers interconnecting multiple blade servers and a dedicated switch module by a fabric topology

Methodology Applied

Scientific EffectOptical transmission: Light

Implementation Method 2

A server chassis includes a plurality of optical transceivers configured to multiplex wavelength-specific optical signals at a plurality of discrete wavelengths

Methodology Applied

Scientific EffectWavelength division multiplexing: Light

Data Source

PatentUS20260082142A1Aggregation of multiplexed optical transceivers in server chassis to establish fabric topology

Publication Date: 2026.03.19 CISCO TECHNOLOGY INC

AI summary

This disclosure describes multiplexed optical transceivers, such as DWDM multiplexer/demultiplexers, which are aggregated in a server chassis to establish a fabric topology interconnecting blade servers to a dedicated switch module. Blade servers installed in the server chassis can utilize not just Ethernet interfaces to connect to network segments, but also PCIe interfaces as well as a combination of Ethernet and PCIe interfaces. The aggregated optical transceivers multiplex and demultiplex wavelength-specific optical signals using a laser source, reducing power consumption over switched fabric ASICs. Servicing of the multiplexed optical transceivers is facilitated by installation and replacement of a laser source. Scaling and redundancy of fabric topology interconnects can be facilitated by selection of laser sources generating expanded ranges of discrete wavelengths. Furthermore, chassis management can be facilitated by configuring network controllers of blade servers to transport chassis management instructions over the fabric topology in-band over a network interface, rather than by an out-of-band pathway.