Link Partner Error Reporting Through FEC Alignment Markers

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional error correction methods, such as FEC processes in high-speed communication networks, fail to provide effective error reporting between link partners, leading to false positives and negatives that degrade network performance and hinder efficient network management.

Innovation Solution

Implementing error correction processes that maintain counters for error events and report error correction information using alignment markers within the FEC architecture, allowing link partners to obtain and generate meaningful link health information for timely network management decisions.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If conventional FEC error correction processes are used in high-speed networks, then error correction capability is provided, but sufficient error reporting between link partners is not achieved

Engineering Contradiction:
Improveerror correction capabilityVSAvoiderror reporting information
Core Design Contradiction:
ReliabilityVSLoss of information

Solution Approach 1:

The patent implements a feedback mechanism where error counters are maintained at both transmitting and receiving ends, and error correction information is reported back to link partners through alignment markers. This allows receiving ends to send detailed error statistics (such as corrected error counts, uncorrectable error counts) to transmitting ends, enabling comprehensive error reporting and link health monitoring that was missing in conventional FEC processes.

Inventive Principle:
Principle #23Feedback

2Measurement precision

If conventional error correction methods are used, then error detection is provided, but false positives and false negatives occur that degrade network performance

Engineering Contradiction:
Improveerror detection accuracyVSAvoidnetwork performance
Core Design Contradiction:
Measurement precisionVSReliability

Solution Approach 1:

The patent replaces simple binary error detection (error/no error) with a detailed statistical measurement system using multiple error counters. Instead of merely detecting whether an error occurred, the system counts and reports specific numbers of corrected errors, uncorrectable errors, and other error metrics. This substitution of mechanical detection with statistical measurement eliminates false positives and negatives by providing quantitative evidence of actual link health.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

3Reliability

If error correction processes are implemented without link partner reporting, then error correction is achieved, but timely network management decisions cannot be made

Engineering Contradiction:
Improveerror correctionVSAvoidnetwork management response time
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

The patent establishes continuous error monitoring and reporting through the alignment marker mechanism. Error counters are continuously updated at both transmitting and receiving ends, and error correction information is reported in every alignment marker transmission. This continuous feedback loop ensures that network management systems always have current error statistics available, enabling timely detection of link degradation and immediate management intervention without waiting for periodic external diagnostics.

Inventive Principle:
Principle #20Continuity of useful action

Data Source

PatentUS11190208B2Techniques for link partner error reporting
Publication Date: 2021.11.30 INTEL CORP
  • US11190208B2 patent drawing
  • US11190208B2 patent drawing
  • US11190208B2 patent drawing

AI summary

Computing devices and techniques for providing link partner health reporting are described. In one embodiment, for example, an apparatus may include at least one memory, and logic, at least a portion of the logic comprised in hardware coupled to the at least one memory, the logic to determine a plurality of error counters, each of the plurality of error counters associated with a number of errors, determine the number of errors for each data unit of a plurality of data units associated with a data block, increment each of the plurality of error counters corresponding with the number of errors for each data unit of the plurality of data units, provide a plurality of error counts for the data block to a link partner, the plurality of error counts corresponding to the number of errors accumulated in each of the plurality of error counters for the data block, and reset the plurality of error counters. Other embodiments are described and claimed.