Sender-Side Message Correctness Checking for Network Fault Tolerance

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

In fault-tolerant computing systems, the existing self-checking technique faces challenges in achieving a prompt reaction time to communication failures as the monitoring entity must intercept and modify message transmission during ongoing communication, which becomes increasingly difficult with higher communication speeds.

Innovation Solution

The method involves a sender monitoring and performing correctness checks on messages and transmission processes, then informing the receiver of any faults after the transmission is complete, allowing the receiver to mark or discard faulty messages, thereby relaxing the reaction time requirement and enabling simpler and faster implementations.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If the monitoring entity intercepts and modifies message transmission during ongoing communication to achieve prompt reaction to failures, then the reliability of the system is improved, but the device complexity and difficulty of implementation increase significantly

Engineering Contradiction:

Improvesystem operabilityVSAvoidimplementation complexity

Core Design Contradiction:

ReliabilityVSDevice complexity

Solution Approach 1:

The sender performs correctness checks on messages before transmission is complete, and prepares correctness information in advance. This allows the monitoring to be done preliminarily without requiring complex interception mechanisms during high-speed transmission, thus improving reliability while reducing implementation complexity

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The monitoring and correctness check functions are extracted from the transmission path and performed separately by the sender before notification. This separates the reliability-critical monitoring function from the high-speed transmission path, reducing the complexity of real-time interception while maintaining system operability

Inventive Principle:

Principle #2Taking out (Extraction)

2Reliability

If the monitoring entity intercepts message transmission during ongoing communication to detect failures promptly, then the reliability is improved, but the communication speed becomes more difficult to maintain

Engineering Contradiction:

Improvefailure detection capabilityVSAvoidcommunication speed

Core Design Contradiction:

ReliabilityVSSpeed

Solution Approach 1:

Correctness checks are performed preliminarily by the sender before the transmission is complete, using pre-computed correctness information. This eliminates the need for complex real-time interception during high-speed transmission, maintaining communication speed while ensuring reliable failure detection

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

Correctness information acts as an intermediary that carries failure detection capability without requiring direct interception of the high-speed message stream. The sender generates this intermediary information separately and notifies the receiver, allowing reliable monitoring without compromising communication speed

Inventive Principle:

Principle #24Intermediary (Mediator)

3Ease of manufacture

If the sender informs the receiver of correctness check results after transmission completion, then the reaction time requirement is relaxed and implementation is simplified, but the time from detection to notification increases

Engineering Contradiction:

Improveimplementation simplicityVSAvoiddetection to notification time

Core Design Contradiction:

Ease of manufactureVSLoss of time

Solution Approach 1:

The sender performs correctness checks preliminarily during the transmission process and prepares correctness information in advance. Although notification occurs after transmission completion, the actual detection happens earlier, reducing the effective detection-to-notification time while maintaining implementation simplicity

Inventive Principle:

Principle #10Preliminary action

Data Source

PatentUS10296407B2Method to detect and to handle failures in the communication in a computer network

Publication Date: 2019.05.21 TTTECH COMPUTERTECHNIK AG

AI summary

A method is provided to detect and handle failures in the communication in a network, including a sender (201, 203) and a receiver (202, 501, 502, 503), where communication between the sender and the receiver is message-oriented. The method includes: (a) the sender sending a message (M101, M101-C, M102-C) to the receiver; (b) the sender monitoring the transmission process of the message inside the sender and/or monitoring the message; (c) the sender executing a correctness check of (i) the message, e.g., its contents, and/or (ii) the transmission process of the message inside the sender; and (d) after the correctness check(s) has/have been completed, the sender informs the receiver of the result of the correctness check(s), wherein (e) the receiver of the message marks the message as being faulty and/or discards the message if the result of a correctness check indicates that the message and/or transmission process is faulty.