Network Availability API for Scalable Router Convergence Measurement

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional network availability measurement techniques face scalability and accuracy issues due to the inability to effectively monitor network convergence in large networks, especially at a millisecond timescale, and lack of completeness in path coverage, which complicates identifying and analyzing network failures.

Innovation Solution

The implementation of a network availability API that utilizes route convergence and traffic demand measurements from routers to compute network availability, providing operational statistics through a clearly organized database and uniform programming interface, allowing for accurate and detailed network performance assessment.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If probe messages are transmitted throughout the network to measure convergence, then network availability can be measured, but the system does not scale well and suffers from accuracy and completeness issues

Engineering Contradiction:
Improvenetwork availability measurement accuracyVSAvoidsystem scalability
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent enables network elements (routers, switches, network devices) to automatically generate and report their own availability metrics without requiring external probe messages. Each network element self-measures its convergence time and availability statistics, eliminating the need for centralized probing infrastructure and achieving both high accuracy and scalability simultaneously

Inventive Principle:
Principle #25Self-service

2Reliability

If probes are injected throughout the network to measure convergence, then availability data can be collected, but completeness of path coverage is insufficient

Engineering Contradiction:
Improveavailability data completenessVSAvoidmeasurement efficiency
Core Design Contradiction:
ReliabilityVSProductivity

Solution Approach 1:

The patent divides the network measurement task into independent segments at each network element. Instead of using centralized probes to cover all paths, each router and switch independently measures its own convergence behavior for its specific forwarding paths. This segmentation ensures complete path coverage while maintaining high measurement efficiency through parallel distributed operations

Inventive Principle:
Principle #1Segmentation

3Adaptability or versatility

If multiple systems are used to provide network availability information, then comprehensive data can be obtained, but applications must interface with multiple systems which complicates implementation

Engineering Contradiction:
Improvenetwork availability data coverageVSAvoidapplication interface complexity
Core Design Contradiction:
Adaptability or versatilityVSEase of operation

Solution Approach 1:

The patent implements a universal standardized interface (API) that provides comprehensive network availability information from multiple network elements through a single unified access point. The API aggregates convergence time statistics, availability metrics, and performance data from various routers and switches, allowing applications to obtain complete network-wide availability information while interfacing with only one standardized system

Inventive Principle:
Principle #6Universality (Multi-functionality)

Data Source

PatentEP2837136B1Network availability analytics
Publication Date: 2019.12.18 CISCO TECHNOLOGY INC
  • EP2837136B1 patent drawingFigure 1
  • EP2837136B1 patent drawingFigure 2
  • EP2837136B1 patent drawingFigure 3

AI summary

In one embodiment, a method includes receiving at a network device, route convergence measurements and traffic demand measurements from a plurality of routers, and computing network availability based on the measurements at the network device. The route convergence measurements are associated with route computations at the routers and the traffic demand measurements include portions of a demand matrix associated with the routers. An apparatus and logic are also disclosed herein.