Adaptive Routing via Backward Notifications in Network Switches

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing communication network routing techniques fail to effectively adapt and reroute packets in response to compromised network links, leading to congestion and potential network bottlenecks, as they often require centralized management and extensive re-routing calculations.

Innovation Solution

The method involves network switches detecting compromised ability to forward packets and sending Adaptive Routing Notifications (ARNs) backwards along the route, allowing preceding switches to modify the route and avoid traversing the compromised switch, with rules based on network topology to ensure semi-local and distributed rerouting.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If centralized management and extensive re-routing calculations are used to handle compromised network links, then routing decisions can be made, but network efficiency deteriorates due to congestion and bottlenecks

Engineering Contradiction:
Improverouting decision accuracyVSAvoidnetwork efficiency
Core Design Contradiction:
ReliabilityVSProductivity

Solution Approach 1:

The patent segments the centralized routing control into distributed autonomous decisions at each network switch. Each switch independently processes AR events and sends notifications only to relevant preceding switches, eliminating the need for centralized management and extensive network-wide recalculations, thus maintaining routing accuracy while improving network efficiency

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent implements a feedback mechanism where network switches monitor their own AR events (congestion, link failures, head-of-line timeouts) and generate notifications that propagate backwards along the route. This feedback loop enables rapid adaptive rerouting without centralized intervention, resolving the contradiction between reliable routing decisions and network efficiency

Inventive Principle:
Principle #23Feedback

2Productivity

If adaptive routing notifications are sent backwards along the route for semi-local rerouting, then network efficiency is improved by avoiding bottlenecks, but device complexity increases due to notification processing rules

Engineering Contradiction:
Improvenetwork efficiencyVSAvoidswitch processing complexity
Core Design Contradiction:
ProductivityVSDevice complexity

Solution Approach 1:

The patent applies local quality by enabling each network switch to autonomously process AR notifications based on its local state and topology knowledge. Each switch independently determines whether to consume or forward notifications using simple rules (e.g., consuming if it's the first switch in the notification chain, forwarding otherwise), avoiding complex centralized processing while maintaining network efficiency

Inventive Principle:
Principle #3Local quality

Solution Approach 2:

The patent inverts the traditional forward-routing approach by sending notifications backwards along the route from the switch detecting the AR event to preceding switches. This inversion enables semi-local rerouting decisions to be made closer to the source, improving network efficiency while keeping individual switch processing relatively simple through well-defined forwarding rules

Inventive Principle:
Principle #13The other way round (Inversion)

3Adaptability or versatility

If distributed rerouting without central entities is implemented, then scalability is improved, but measurement precision deteriorates in detecting compromised links

Engineering Contradiction:
Improvenetwork scalabilityVSAvoidAR event detection accuracy
Core Design Contradiction:
Adaptability or versatilityVSMeasurement precision

Solution Approach 1:

The patent implements self-service by enabling each network switch to autonomously detect AR events (congestion, link failures, head-of-line timeouts) and process notifications without external assistance. Each switch monitors its own state and makes independent rerouting decisions, achieving both high scalability and accurate event detection through distributed intelligence rather than centralized measurement

Inventive Principle:
Principle #25Self-service

Data Source

PatentUS9634940B2Adaptive routing using inter-switch notifications
Publication Date: 2017.04.25 MELLANOX TECHNOLOGIES LTD(IL)
  • US9634940B2 patent drawing
  • US9634940B2 patent drawing
  • US9634940B2 patent drawing

AI summary

A method includes receiving in a network switch of a communication network communication traffic that originates from a source node and arrives over a route through the communication network traversing one or more preceding network switches, for forwarding to a destination node. In response to detecting in the network switch a compromised ability to forward the communication traffic to the destination node, a notification is sent to the preceding network switches. The notification is to be consumed by the preceding network switches and requests the preceding network switches to modify the route so as not to traverse the network switch.