Fault Tolerance Processing for Data Access Systems

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing data processing systems often restart due to invalid data returned from storage mediums, as they lack effective abnormality checks, leading to inefficiencies and potential data loss.

Innovation Solution

A fault tolerance processing method is introduced, where an internal controller generates a response status identifier based on indication information from an external controller, allowing the processor to determine the validity of data before performing operations, thereby preventing system restarts.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If the processor directly processes data returned from storage medium without abnormality check, then the data processing speed is fast, but the system reliability deteriorates due to potential invalid data causing system restart

Engineering Contradiction:
Improvesystem reliabilityVSAvoiddata processing complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent applies preliminary action by performing abnormality detection on returned data before the processor executes operations. The detection unit checks whether the data is valid (e.g., whether ECC correction succeeded) before the processor consumes the data, preventing invalid data from causing system restart while allowing normal processing to proceed without interruption.

Inventive Principle:
Principle #10Preliminary action

2Reliability

If the system performs abnormality detection and fault tolerance processing on returned data, then the system stability is improved, but the data processing time increases

Engineering Contradiction:
Improvesystem stabilityVSAvoiddata processing time
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

The patent maintains continuity of useful action by performing detection in parallel with processing or immediately upon data return, rather than interrupting the processing flow. The detection unit operates on returned data without blocking the processor, and only intervenes when actual abnormalities are detected, thus minimizing time loss while maintaining system stability.

Inventive Principle:
Principle #20Continuity of useful action

3Productivity

If the processor consumes data without verifying its validity, then the processing efficiency is high, but the data integrity deteriorates leading to potential data loss

Engineering Contradiction:
Improveprocessing efficiencyVSAvoiddata integrity
Core Design Contradiction:
ProductivityVSLoss of information

Solution Approach 1:

The patent implements feedback by having the detection unit continuously monitor the validity of data returned from the storage medium and provide status information to the processor. The processor receives feedback about data validity (e.g., ECC correction status) and adjusts its processing accordingly, ensuring data integrity while maintaining high processing efficiency for valid data.

Inventive Principle:
Principle #23Feedback

Data Source

PatentEP3796615B1Fault tolerance processing method, device, and server
Publication Date: 2023.06.07 HUAWEI TECH CO LTD
  • EP3796615B1 patent drawingFigure 1
  • EP3796615B1 patent drawingFigure 2~3
  • EP3796615B1 patent drawingFigure 4

AI summary

This application provides a fault tolerance processing method, and the method includes: sending, by a server, a data read packet to an external controller, where the data read packet is used to request the external controller to read data in an indicated storage location, and a packet format of the data read packet complies with a packet format specified in the memory-semantic network protocol; receiving, by the server, indication information sent by the external controller, where the indication information is used to indicate a response status of the external controller to the data read packet; and determining, by the server, the response status of the external controller to the data read packet based on the indication information, and performing fault tolerance processing when the response status is an abnormal response. This method can effectively reduce a probability of a restart of a data access system due to invalidity.